Graves glanced at Phelps. Phelps was watching them all intently, pretending he was following the discussion. Graves knew that it was over Phelps' head.

`Just before I finished,' Decker said, `we discovered that an unauthorized station was tapping into the system. We called it Sigma Station, but we were unable to characterize it. By that I mean that we knew Sigma was drawing information, but we didn't know where, or how.'

He flipped to a green sheet of computer printout and pushed it across the table to Graves. `Sigma is the underlined station. You can see that on this particular day, July 21st, 1972, it tapped into the system at ten oh four rnt Eastern time and maintained the contact for seven minutes; then it broke out. We determined that Sigma was tapping in at around ten o'clock two or three nights a week. But that was all we knew.'

Decker turned to Venn, who said, `I came into the picture at this point. I'd been at Bell Labs working on telephone tracer mechanisms. The telephone company has a problem with unauthorized calls - calls verbally charged to a phone number, calls charged to a wrong credit card number, that kind of thing. I was working on a computer tracing system. Defence asked me to look at the Sigma Station problem.'

`One ought to say,' Phelps said, `that the data bank being tapped by Sigma was a Defence bank.'

`Yes,' Venn said. `It was a Defence bank. With two or three taps a week at about ten PM. That was all I knew when I began. However, I made some simple assumptions. First, you've got to have a computer terminal in order to tap the system. That is, once you've called the number that links you to the computer, you must use a teletypewriting or CRT apparatus compatible with the Defence system.'

`Are those terminals common?'

`No,' Venn said. `They are quite advanced and fairly uncommon. I started with a list of them.'