In Natural evolution, a mutation may be lethal, causing an abrupt end to an evolving lineage. This fact has a tendency to cause evolution to "prefer" mutationally robust solutions (which can in turn slow innovation), an effect that has been studied previously, especially in the context of evolution on neutral plateaux. Here, we tackle related issues but from the perspective of a practical optimization scenario. We wish to evolve a finite population of entities quickly (i.e. improve them), but when a lethal solution (modelled here as one below a certain fitness threshold) is evaluated, it is immediately removed from the population and the population size is reduced by one. This models certain closed-loop evolution scenarios that may be encountered, for example, when evolving nano-technologies or autonomous robots. We motivate this scenario, and find that evolutionary search performs best in a lethal environment when limiting randomness in the solution generation process, e.g. by using elitism, above-average selection pressure, a less random mutating operator, and no or little crossover. For NKα landscapes, these strategies turn out to be particularly important on rugged and non-homogeneous landscapes (i.e. for large K and α).