Revisiting the Mousetraps Experiment: Not Just about Nuclear Chain Reactions

Abstract

We present here the first quantitative measurements of a classic experiment, that of the “mousetrap chain reaction”. It was proposed for the first time in 1947 to illustrate the chain reaction occurring in nuclear fission. It involves several spring-loaded mousetraps loaded with solid balls. Once one trap is made to snap, it releases two balls that may trigger the other traps. The result is a chain reaction that rapidly flares and then subsides as most traps have been triggered. The experiment has been popular as a scientific demonstration, but it does not seem that quantitative data were ever reported about it, nor that it was described using a model. We set out to do exactly that, and we can report for the first time that the mousetrap experiment can be fitted by a simple dynamic model derived from the well-known Lotka-Volterra one. We also discuss the significance of this experiment beyond nuclear chain reactions, providing insight into a variety of fields (chemistry, biology, memetic, natural resources exploitation) involving complex adaptive systems.