The code protein has two functions, and another function which is much more important than being a code protein. The code protein binds to the RNA and stops the translation of the enzyme. So only during the first ten, twelve minutes you produce enzyme, and that means specific replicase, and after that it is stopped. Now what happens then? These produced replicases, these enzymes which can replicate now the virus RNA, now start to compete with the ribozymes in a hypercyclic fashion. In other words they make new RNA, the new RNA makes new protein and vice versa, and only those proteins feed back which are made from these.

Furthermore, you might ask why does the RNA production stop after that time. Well, it's the error rate. After that type, enough errors have accumulated that making even more enzyme now would make two erroneous ones and that would interfere with the real process. But during that short time between ten, twelve minutes, you now get a hyperbolic type of growth because at the same time it's not only replication which would make an exponential law, but at the same time the number of enzyme molecules increase also. So it's not only that you get two molecules, four, eight, sixteen RNA molecules; at the same time you make more and more enzyme and that's exactly what we calculated for a hypercycle until the enzyme molecules all are saturated. That's like a titration, then the reaction stops, and then with constant rate, according to the saturated enzymes, it produces for the rest of the time enough RNA molecules, enough proteins, in a very controlled way so that you have at the end for each RNA molecule the sufficient number of code proteins, so that you can form complete virus particles. Think... you make more nucleic acid than protein to code, then you get incompletely coded virus particles which would be immediately degraded, and so on. So it's a highly regulated system with promotion and inhibition, and the feedback is that you have a specific replicase which only replicates its virus RNA and doesn't touch any host RNA.

So we looked at this model, we calculated from the experiment and we found exactly the hypercycle fulfilled there. In other words, only a hypercyclic law, replication law, yields your hyperbolic growth laws, all the others would yield exponential laws and we could definitely prove that the law is not exponential but rather hyperbolic. In other words, it very slowly starts and suddenly almost goes to infinite, but it can't of course go to infinite, but a hyperbola has a singularity and that means where the goals would go to infinite. So hypercycle is a proven system.