A couple more points. Depending on how diverse the population is, a high mutation rate may not help when selecting the fittest children. There is also a chance that you will find a local maxima, and not the best solution overall. Introducing a concept of multiple "species" to the population will sometimes help finding other solutions outside the local maxima, but then this complicates the crossover phase. If the "genes" are of equal length and therefore closely related species, you may be able to generate some offspring in the crossover phase, but then you are usually introducing another species at that time. Depending on the solution you are trying to find, this may or may not be appropriate.