Experiments that go according to plan can be useful. But the biggest scientific advances often emerge from those that do not. Such is the case with a study just reported in the Proceedings of the National Academy of Sciences. When they began it, Hector DeLuca of the University of Wisconsin, Madison, and his colleagues had been intending to examine the effects of ultraviolet (UV) light on mice suffering from a rodent version of multiple sclerosis (MS). By the project's end, however, they had in their hands two substances which may prove valuable drugs against the illness. Multiple sclerosis is an autoimmune disease. This means it is caused by a victim's immune system turning on and destroying parts of his own body. In the case of MS the targets of these attacks, which may continue for years, are the fatty sheaths that insulate nerve cells and thus help nervous impulses to propagate. People suffering from MS are often weakened, and sometimes physically disabled by it, and may also become blind. What drives the immune system to behave in this way remains mysterious, but in the 1970s researchers uncovered a promising clue when they noticed that MS is rarer near the equator than it is at high latitudes. The first hypothesis proposed to explain this observation was that vitamin D (a substance created by sunlight's action on precursor molecules in the skin) might be helping to prevent MS. That made sense, since those living in the tropics receive more sunlight than those in temperate zones. Sadly, follow-up experiments failed to support the notion. Those experiments did, though, lead Dr. DeLuca to discover that the preventive effect is associated with a particular sort of sunlight—UV with a wavelength of between 300 and 315 nanometres (billionths of a metre). His latest experiment was intended to dig deeper into this observation, by using this type of light to irradiate mice that had been injected with chemicals known to cause the rodent equivalent of MS.