This statement from Heywood for me is profound: "It also suggests that cycle by cycle variations in the position and growth rate of the flame kernel at very early times are the very cause of the cycle by cycle fluctuations in the pressure curves for spark ignition engines."