In recent years, the following ideas have been expressed: (a) that all cases of a discrete, inherited neuromuscular syndrome should prove to be due to a single biochemical defect, (b) that any single biochemical defect should give rise only to one syndrome, and (c) that an enzymatic defect cannot give rise to a disease unless there is virtual absence of activity, that is, less than 5% or 10% of the normal value. We review evidence from research in neuromuscular, neurological, and other genetic diseases of humans that suggest the contrary. There are now examples of single clinical syndromes related to each of several defects, of defects of one biochemical reaction related to two or more distinct clinical syndromes, and of partial defects associated with disease in a way that suggests a causal relationship.