Anders said, "Listen, what does epilepsy have to do with cardiac pacemakers?"

"Nothing," she said, "except that Benson has a brain pacemaker, very similar to a cardiac pacemaker."

Anders flipped open his notebook. "You better start from the beginning," he said, "and go slowly."

"All right." She set down her drink. "Let me make one call first."

Anders nodded, sat back, and waited while she called McPherson. Then, as calmly as she could, she explained everything she knew to the policeman.

McPherson hung up the telephone and stared out his office window at the morning sun. It was no longer pale and cold; there was the full warmth of morning. "That was Ross," he said.

Morris nodded from the corner. "And?"

"Benson came to her apartment. She lost him."

Morris sighed.

"It doesn't seem to be our day," McPherson said. He shook his head, not taking his eyes off the sun. "I don't believe in luck," he said. He turned to Morris. "Do you?"

Morris was tired; he hadn't really been listening. "Do I what?"

"Believe in luck."

"Sure. All surgeons believe in luck."

"I don't believe in luck," McPherson repeated, "Never did.

I always believed in planning." He gestured to the charts on his wall, then lapsed, staring at them.

The charts were large things, four feet across, and intricately done in many colors. They were really glorified flow charts with timetables for technical advances. He had always been proud of them. For instance, in 1967 he had examined the state of three areas - diagnostic conceptualization, surgical technology, and microelectronics - and concluded that they would all come together to allow an operation for psychomotor epilepsy in July of 1971. They had beaten his estimate by four months, but it was still damned accurate.

"Damned accurate," he said.

"What?" Morris said.

McPherson shook his head. "Are you tired?"

"Yes."

"I guess we're all tired. Where's Ellis?"

"Making coffee."

McPherson nodded. Coffee would be good. He rubbed his eyes, wondering when he would be able to sleep. Not for a while - not until they had Benson back. And that could take many hours more, perhaps another day.

He looked again at the charts. Everything had been going so well. Electrode implantation four months ahead of schedule. Computer simulation of behavior almost nine months ahead - but that, too, was having problems. George and Martha programs were behaving erratically. And Form Q?

He shook his head. Form Q might never get off the ground now, although it was his favorite project, and had always been. Form Q on the flow chart for 1979, with human application beginning in 1986. In 1986 he would be seventy-five years old - if he was still alive - but he didn't worry about that. It was the idea, the simple idea, that intrigued him.

Form Q was the logical outgrowth of all the work at the NPS. It began as a project called Form Quixoticus, because it seemed so impossible. But McPherson felt certain that it would happen because it was so necessary. For one thing, it was a question of size; for another, a question of expense.

A modern electronic computer - say, a third-generation IBM digital computer - would cost several million dollars. It drew an enormous amount of power. It consumed space voraciously. Yet the largest computer still had the same number of circuits as the brain of an ant. To make a computer with the capacity of a human brain would require a huge skyscraper. Its energy demands would be the equivalent of a city of half a million.

Obviously, nobody would ever try to build such a computer using current technology. New methods would have to be found - and there wasn't much doubt in McPherson's mind what the methods would be.

Living tissues.

The theory was simple enough. A computer, like a human brain, was composed of functioning units - little flip-flop cells of one kind or another. The size of those units had shrunk considerably over the years. They would continue to shrink as large-scale integration and other microelectronic techniques improved. Power requirements would also decrease. But the individual units would never become as small as a nerve cell, a neuron. You could pack a billion nerve cells into one cubic inch. No human miniaturization method would ever achieve that economy of space. Nor would any human method ever produce a unit that operated on so little power as a nerve cell.

Therefore, make your computers from living nerve cells. It was already possible to grow isolated nerve cells in tissue culture. It was possible to alter them artificially in different ways. In the future, it would be possible to grow them to specification, to make them link up in specified ways.

Once you could do that, you could make a computer that was, say, six cubic feet in volume, but contained thousands of billions of nerve cells. Its energy requirements would not be excessive; its heat production and waste products would be manageable. Yet it would be the most intelligent entity on the planet, by far.

Form Q.

Preliminary work was already being done in a number of laboratories and government research units around the country. Advances were being made.

But for McPherson the most exciting prospect was not a superintelligent organic computer. That was just a side product. What was really interesting was the idea of an organic prosthesis for the human brain.

Because once you developed an organic computer - a computer composed of living cells, and deriving energy from oxygenated, nutrified blood - then you could transplant it into a human being. And you would have a man with two brains.

What would that be like? McPherson could hardly imagine it. There were endless problems, of course. Problems of interconnection, problems of location, speculative problems about competition between the old brain and the new transplant. But there was plenty of time to solve that before 1986. After all, in 1950 most people still laughed at the idea of going to the moon.

Form Q. It was only a vision now, but with funding it would happen. And he had been convinced that it would happen, until Benson left the hospital. That changed everything.

Ellis stuck his head in the office door. "Anybody want coffee?"

"Yes," McPherson said. He looked over at Mor.

"No," Morris said. He got up out of his chair. "I think

I'll replay some of Benson's interview tapes."

"Good idea," McPherson said, though he did not really think so. He realized that Morris had to keep busy - had to do something, anything, just to remain active.

Morris left, Ellis left, and McPherson was alone with his multicolored charts, and his thoughts.

It was noon when Ross finished with Anders, and she was tired. The Scotch had calmed her, but it had intensified her fatigue. Toward the end she had found herself stumbling over words, losing track of her thoughts, making statements and then amending them because they were not exactly what she had intended to say. She had never felt so tired, so drugged with fatigue, in her life.

Anders, on the other hand, was maddeningly alert. He said,

"Where do you suppose Benson is now? Where would he be likely to go?"

She shook her head. "It's impossible to know. He's in a post-seizure state - post-ictal, we call it - and that's not predictable."

"You're his psychiatrist," Anders said. "You must know a lot about him. Isn't there any way to predict how he'll act?"

"No," she said. God, she was tired. Why couldn't he understand? "Benson is in an abnormal state. He's nearly psychotic, he's confused, he's receiving stimulations frequently, he's having seizures frequently. He could do anything."

"If he's confused…" Anders let his voice trail off.

"What would he do if he was confused? How would he behave?"

"Look," she said, "it's no good. It's no damned good, working that way. He could do anything."

"Okay," Anders said. He glanced at her briefly, and sipped his coffee.

Why couldn't he just let it go, for Christ's sake? His desire to psych out Benson and track him down was ludicrously unrealistic. Besides, everybody knew how this was going to turn out. Somebody would spot Benson and shoot him, and that would be the end of it. Even Benson had said-

She paused, frowning. What had he said? Something about how it would all end. What were his exact words? She tried to remember, but couldn't. She had been too frightened to pay close attention.

"These are the impossible ones," Anders said, getting up and walking to the window. "In another city, you might have a chance, but not in Los Angeles. Not in five hundred square miles of city. It's bigger than New York, Chicago, San Francisco, and Philadelphia put together. Did you know that?"

"No," she said, hardly listening.

"Too many places to hide," he said. "Too many ways to escape - too many roads, too many airports, too many marinas. If he's smart, he's left already. Gone to Mexico or to Canada."

"He won't do that," she said.

"What will he do?"

"He'll come back to the hospital."

There was a pause. "I thought you couldn't predict his behavior," Anders said.

"It's just a feeling," she said, "that's all."

"We'd better go to the hospital," he said.

The NPS looked like the planning room for a war. All patient visits had been canceled until Monday; no one but staff and police were admitted to the fourth floor. But for some reason, all the Development people were there, and they were running around with horrified looks on their faces, obviously worried that their grants and their jobs were in jeopardy. Phones rang constantly; reporters were calling in; McPherson was locked in his office with hospital administrators; Ellis was swearing at anyone who came within ten yards of him; Morris was off somewhere and couldn't be found; Gerhard and Richards were trying to free some telephone lines so they could run a projection program using another computer, but all the lines were in use.

Physically, the NPS was a shambles - ashtrays heaped with cigarette butts, coffee cups crumpled on the floor, half-eaten hamburgers and tacos everywhere, jackets and uniforms thrown across the backs of chairs. And the telephones never stopped ringing: as soon as anyone hung up on a call, the phone rang again instantly.