 The third model is called nucleation condensation and the idea is not as complicated as it sounds. The idea here is that we start from a chain and this chain collapses but let's see if we can find something halfway between the two previous models. So instead of saying that it collapses completely, let's assume that there are a few critical contacts formed somewhere in some sort of core here, a nucleus, just as when you freeze ice, right, that when you freeze water into ice there has to be a first grain of ice crystals somewhere, the nucleus and after that you would have more water molecules condensing on that and grow the ice part. So the idea here is that this would be a fast process and then gradually we would have all the secondary structure elements grow from this nucleus until we had formed the entire protein. This actually agrees quite well with some of the things we've seen, that there is a bit of structure here but it's not perfect. There will have to be some sort of transition but right now I can neither prove nor disprove this model, at least it's not disproven. To show that it's correct we're going to need to look a little bit more at kinetics, how fast processes happen and in particular I'm going to need tools that can connect experimental things that I can observe that is for instance how fast something falls but use that to draw conclusions about this state because the problem is that this is what type of state is it, it's a transition state and I can't measure transition state, I can't observe them directly. Something like the diffusion collision model that alpha helix and beta sheet might be in local free energy minimum, a stable part, this is likely going to be a local free energy maximum which complicates things a bit. So to understand this we're going to need to go back to chemical reaction kinetics but now study protein folding or in particular processes like this in more detail.