I had been working at Columbia University, I had a number of friends working on molecular beams, and I knew all about molecular beams as well as the properties of molecules. So I pulled out a piece of paper in my pocket, it was an envelope, and started working with the numbers to see how many molecules would one need in order to produce enough energy so it would be useful, and how can you get that many molecules. And I realized one could send a beam of molecules in a vacuum, have an electric field which pulled out the ones you didn’t want, left the ones you did want going straight along. And they’d come into a cavity, and as they entered the cavity, the waves could be bouncing back and forth in the cavity, and take the energy out of the molecules. Now the cavity was something I learned about from microwaves. At Bell Laboratories, I had worked with cavities. The beam was something I had learned about at Columbia University. I knew a lot about that because my friends were working on that. Molecules, of course, I had been working with at Columbia University. So all of these things one puts together, and suddenly I realized, now wait a minute, that can do it. And I showed with calculations that, yes, one can get enough energy to make it work. And of course I was exhilarated by the idea that, yes, it looks like it could work. It was marginal. I said, “Well, it’s going to be difficult, but I believe it can be done.”