 The photoelectric effect is where you shine light on a metal and it emits electrons. That's how we understand it today. When it was first discovered we didn't even know what an electron was. So in early experimental exploration of the photoelectric effect it was just noticed that light, in particular ultraviolet light, seemed to have some strong effect on whether metals would emit electrical charge or not. In 1887 Heineck Hertz noticed that a spark was more likely to jump the gap between two electrodes if it was illuminated by ultraviolet light. A flow of investigations followed and it was found that the amount of current depended on a lot of things. It depended on the metal type, it depended on the charge on the metal, it depended on the atmosphere, it depended on the oxidation level, how well polished the surface was, and all sorts of things like that. Alexander Stolatov did a very detailed quantitative analysis of the photoelectric effect and he discovered the direct proportionality between how much light, the intensity of light that was incident on the metal, and the amount of current. In 1900-1902 Philip Leonard was investigating the same kind of process in the ionization of gases by ultraviolet light and he noticed that the energy of the individual charges that were being emitted by the gas increased with the frequency of the light. It turns out that these two experimental facts are a bit surprising to people who are modelling light using Maxwell's theory of electromagnetism. The power in a light beam is proportional to the intensity of that light beam and the frequency of that light beam. And so you'd expect that the power delivered by the light would go up with both of those quantities. So why is the current going up as you imply more intensity? Well, you'd expect more energy to be delivered so you might expect that over time more charged particles, let's just call them electrons now, might be emitted. But why would the energy of those emitted charges only depend on the frequency? So the energy of the emitted charges we might expect to also depend on the intensity, but it does not. And if very weak light is incident on the metal, then you expect that eventually enough energy would be put into that metal to emit one of these electrons. Even if there was some kind of binding energy you had to break, you'd expect eventually you'd have enough energy from the light building up to break it free. And what that would lead to, it would mean that if you had a very weak dim light, then there'd be some kind of time delay from when you turned on the light and when the charge came off. But it turns out that's not how it works. There seems to be no measurable time delay between when the light is incident and when the charges come off. And yet there's this very strange frequency dependence on the energy and lack of intensity dependence on the energy.