Though theoretical biologists have responded enthusiastically to the challenge of resolving the paradox of sex, there has not been an equivalent amount of experimental testing of the hypotheses they have put forward. This paper briefly reviews the essential elements of the models proposed to explain the predominance of sexual reproduction, and describes the existing experimental evidence. Some of the experimental approaches have involved tests of the models' assumptions; work done to date, for example, suggests that the theoretically twofold 'cost of sex' may be an over-estimate for animals, but an under-estimate for plants [see also the chapter by Lewis-Ed.]. The assumption that truncation selection occurs in nature has found no experimental support. Other experiments have tested the models' predictions. There is little evidence that sibling competition is mitigated by genetic diversity among progeny. Several studies have provided evidence that minority genotypes have higher fitness, but generally fail to reveal a mechanism for that advantage. Overall, no one model emerges as a likely single 'cause' for the maintenance of sexual reproduction. Future progress is likely to come, not from further tests of the models' predictions (which are necessary logical outcomes of particular sets of assumptions), but rather from examining the frequency with which their assumptions are met in nature.