Hunt sighed and pointed a weary finger toward the screen.

"I'm not saying it, Chris," he reminded the professor. "The numbers are. There are the facts--check'em." Hunt leaned forward and cocked his head to one side, at the same time contorting his features into a frown as if he had just been struck with a sudden thought. "What were you saying a minute ago about people wanting to fit the evidence to suit the answers they'd already made their minds up about?" he asked.

At the age of eleven, Victor Hunt had moved from the bedlam of his family home in the East End of London and gone to live with an uncle and aunt in Worcester. His uncle--the odd man out in the Hunt family--was a design engineer at the nearby laboratories of a leading computer manufacturer and it was his patient guidance that first opened the boy's eyes to the excitement and mystery of the world of electronics.

Some time later young Victor put his newfound fascination with the laws of formal logic and the techniques of logic-circuit design to its first practical test. He designed and built a hard-wired special-purpose processor which, when given any date after the adoption of the Gregorian calendar in 1582, would output a number from 1 to 7 denoting the day of the week on which it had fallen. When, breathless with expectation, he switched it on for the first time, the system remained dead. It turned out that he had connected an electrolytic capacitor the wrong way around and shorted out the power supply.

This exercise taught him two things: Most problems have simple solutions once somebody looks at things the right way, and the exhilaration of winning in the end makes all the effort worthwhile. It also served to reinforce his intuitive understanding that the only sure way to prove or disprove what looked like a good idea was to find some way to test it. As his subsequent career led him from electronics to mathematical physics and thence to nucleonics, these fundamentals became the foundations of his permanent mental makeup. In nearly thirty years he had never lost his addiction to the final minutes of mounting suspense that came when the crucial experiment had been prepared and the moment of truth was approaching.

He experienced that same feeling now, as he watched Vincent Carizan make a few last-minute adjustments to the power-amplifier settings. The attraction in the main electronics lab at Pithead Base that morning was an item of equipment recovered from the Ganymean ship. It was roughly cylindrical, about the size of an oil drum, and appeared to be rather simple in function in that it possessed few input and output connections; apparently it was a self-contained device of some sort, rather than a component in some larger and more complex system.

However, its function was far from obvious. The engineers at Pithead had concluded that the connections were intended as power inlet points. From an analysis of the insulating materials used, the voltage clamping and protection circuits, the smoothing circuits, and the filtering arrangements, they had deduced the kind of electrical supply it was designed to work from. This had enabled them to set up a suitable arrangement of transformers and frequency converters. Today was the day they intended to switch it on to see what happened.

Besides Hunt and Carizan, two other engineers were present in the laboratory to supervise the measuring instruments that had been assembled for the experiment. Frank Towers observed Canzan's nod of satisfaction as he stepped back from the amplifier panel and asked:

"All set for overload check?"

"Yep," Carizan answered. "Give it a zap." Towers threw a switch on another panel. A sharp clunk sounded instantly as a circuit breaker dropped out somewhere in the equipment cabinet behind the panel.

Sam Mullen, standing by an instrumentation console to one side of the room, briefly consulted one of his readout screens. "Current trip's functioning okay," he announced.

"Unshort it and throw in some volts," Carizan said to Towers, who changed a couple of control settings, threw the switch again and looked over at Mullen.

"Limiting at fifty," Mullen said. "Check?"

"Check," Towers returned.

Carizan looked at Hunt. "All set to go, Vic. We'll try an initial run with current limiters in circuit, but whatever happens our stuff's protected. Last chance to change your bet; the book's closing."

"I still say it makes music." Hunt grinned. "It's an electric barrel organ. Give it some juice."

"Computers?" Carizan cocked an eye at Mullen.

"Running. All data channels checking normal."

"Okay then." Carizan rubbed the palms of his hands together. "Now for the star turn. Live this time, Frank--phase one of the schedule."

A tense silence descended as Towers reset his controls and threw the main switch again. The readings on the numeric displays built into his panel changed immediately.

"Live," he confirmed. "It's taking power. Current is up to the maximum set on the limiters. Looks like it wants more." All eyes turned toward Mullen, who was scanning the computer output screens intently. He shook his head without looking around.

"Nix. Makes a dodo look a real ball of fire."

The accelerometers, fixed to the outside of the Ganymean device standing bolted in its steel restraining frame on rubber vibration absorbers, were not sensing any internal mechanical motion. The sensitive microphones attached to its casing were picking up nothing in the audible or ultrasonic ranges. The heat sensors, radiation detectors, electromagnetic probes, gaussmeters, scintillation counters, and variable antennas--all had nothing to report. Towers varied the supply frequency over a trial range but it soon became apparent that nothing was going to change. Hunt walked over to stand beside Mullen and inspect the computer outputs, but said nothing.

"Looks like we need to wind the wick up a little," Carizan commented. "Phase two, Frank." Towers stepped up the input voltage. A row of numbers appeared on one of Mullen's screens.

"Something on channel seven," he informed them. "Acoustic." He keyed a short sequence of commands into the console keyboard and peered at the wave form that appeared on an auxiliary display. "Periodic wave with severe even-harmonic distortion, low amplitude . . . fundamental frequency is about seventy-two hertz."

"That's the supply frequency," Hunt murmured. "Probably just a resonance somewhere. Shouldn't think it means much. Anything else?"

"Nope."

"Wind it up again, Frank," Carizan said.

As the test progressed they became more cautious and increased the number of variations tried at each step. Eventually the characteristics of the input supply told them that the device was saturating and seemed to be running at its design levels. By this time it was taking a considerable amount of power but apart from reporting continued mild acoustic resonances and a slight heating of some parts of the casing, the measuring instruments remained obstinately quiet. As the first hour passed, Hunt and the three UNSA engineers resigned themselves to a longer and much more detailed examination of the object, one that would no doubt involve dismantling it. But, like Napoleon, they took the view that lucky people tend to be people who give luck a chance to happen; it had been worth a try.

The disturbance generated by the Ganymean device was, however, not of a nature that any of their instruments had been designed to detect. A series of spherical wave fronts of intense but highly localized space-time distortion expanded outward from Pithead Base at the speed of light, propagating across the Solar System.

Seven hundred miles to the south, seismic monitors at Ganymede Main Base went wild and the data validation programs running in the logging computer aborted to signal a system malfunction.