Now, a friend of mine, George Placzek, who drew my attention to the problem, he and I applied all this to the Selection Principle. Which vibration of a molecule will appear in the Raman effect and how will it appear? And again, I can best illustrate what we have done by a concrete example, one of my favorite molecules, carbon dioxide. It's simple enough so you can readily understand what you talk about. It still has the complexity to be a little less boring than the simplest examples. Here is a carbon dioxide molecule: carbon, two oxygens, on a straight line. Now consider two vibrations. This vibration, the carbon remains in the same position, the oxygens move up and dow- away and toward. And this vibration. The carbon moves up and down, relative to the two oxygens. If it- you give it a moment of thought, you will see that purely classically, what I describe to you as Raman effect, should behave quite differently in the two cases. Here, you might go from an easily polarized state to a less easily polarized state, to an easily polarized state, to a less easily polarized state. The polarizability changes, goes through a complete cycle. When the molecular vibration goes through a complete cycle, that's a simple thing that you would expect. And that corresponds to the case where, in the Raman effect, the vibration, this one, will be excited with a single quantum. But now, look at this. The carbon vibrates against the two oxygen atoms, oxygen atoms. Let the carbon start at the center, has a polarizability. It gets off the center and now it has a greater polarizability. It gets back to the center, it has the original polarizability. If it gets off the center, a greater polarizability, back to the original center, original polarizability. In a complete vibration, from here to here to here to here, the polarizability had two maxima, here and here, and two minima, when the carbon went through the central position. You will never see one quantum added to the original frequency of light, but only two quanta. Because half a vibration is enough to re-establish the original position- original polarizability. And therefore, the polarizability will change twice as many times per second. Then the complete vibration changes. This thing, this kind of idea, applied to vibrations and incidentally, which is the main thing Placzek and I did, applied to rotations - the rotational frequencies also can be imposed on the original frequency. That is what we did for a few years, few- quite a few months, at any rate. Almost all the two years I have been in Göttingen.