Accidental scientists.

Goof-ups, wacky results, and plain bad luck can ruin a science project--or turn a humdrum experiment into a humdinger.

Your experiment is almost over and you can already see yourself making your mark on science. Then the unexpected happens. Instead of beans, a strange fungus springs up in your pots. A cockroach climbs into your test tube and turns the solution brown. Your data make your hypothesis look like Bart Simpson's homework!

Throw up your hands in despair? Na-a-a-ah! You just have to know what to make of your mistakes. Here are the stories of three people who did.

STAR STRUCK

Jocelyn Bell didn't botch her experiment, but her results left her puzzled--until she saw the light (a pulsing radio beam, actually) from a new kind of star.

At the age of 24, Bell had traveled to England from Ireland to work with a team of astronomers at Cambridge University. They were studying radio waves coming from stars. (Stars give off energy not only as light but also as radio waves.)

Bell's job was to pore over records of radio signals picked up by antennas and look for variations in the strength of the signal. Bell expected to see the height of the signals vary during the day: The theory was that our Sun somehow interferes with signals coming from faraway stars. So when Bell noticed higher peaks on a readout from the middle of the night, she thought it seemed bizarre. Then, like any good scientist, she set herself the task of figuring out what was going on.

By adjusting the instruments, Bell was able to spread the signal out. It was "a bit like doing a photographic enlargement," she says. It turned out that these were no ordinary fluctuations. The radio waves were chopped into rock-steady pulses, like a radio heartbeat coming from one part of the sky. This was very unstarlike behavior.

Bell considered the possibilities. Were the antennas broken? Was the signal coming from a statellite, perhaps? Or were extraterrestrial aliens trying to make contact?

It took months of tuning in and poring over data for Bell to convince herself that these explanations didn't add up. The only explanation left was the most exciting of all: Bell realized she'd discovered a whole new type of star. She gave it the name pulsar.

Twenty-five years later, astronomers are still puzzlying over what makes pulsars pulse. Maybe in the next 25 years someone--you?--will find out.

THINK TANK

When it rained on Percy Julian's experiment, he made the best of it.

Julian was struggling to come up with cheap methods of making drugs called steroids. You may think of steroids as dangerous drugs athletes take to build muscle, but steroids include some important medicines as well. Best known is cortisone, which doctors use to heal skin rashes, soothe injuries, and treat crippling rheumatoid arthritis.

When Julian began his work in the 1930s, steroid medicines were incredibly expensive; they had to be extracted in tiny amounts from the organs of cows. Julian and other chemists wanted to synthesize steroids in the laboratory, that is, build them artificially.

Julian had already made a name for himself by being the first to synthesize another expensive drug. He'd done it by combining molecules found in cheap soybean oil. An African-American scientist at a time of open racism, he had met an extra challenge by beating some prestigious English chemists who were also racing to synthesize that drug.

In the long and complicated process of breaking soybean oil into its chemical building blocks, Julian isolated the building blocks he needed and disregarded the rest. One of the unused by-products was a white crystal called stigmasterol, which Julian thought little about ... until an accident a few years later brought it back to mind.

Julian was now the head scientist at a soybean oil company. One morning, a worker rushed into his office to tell him that rain was leaking into a huge tank of purified soybean oil. It sounded like a lot of Julian's raw material would be ruined--which is exactly what happened. But what Julian got in its place turned out to be more important.

When he peered into the tank, Julian saw a white crystalline substance. It looked awfully familiar--like one of those by-products he used to throw away. Could it really be stigmasterol? You bet!

That was big news because, by then, Julian had learned that stigmasterol was the key building block needed to make steroids. And now an easy way to make that building block had spilled into his hands: no more complicated chemistry--just add water. The water reacts with stigmasterol in the soybean oil to form the crystalline solid.

Within a few years, the company Julian worked for was producing steroids so inexpensively that anyone could afford to be treated with them.

BLUNDER TO WONDER DRUG

One young Scottish doctor by the name of Alexander Fleming flubbed a science project and wound up with the miracle drug of the 20th century: penicillin.

In 1928, Fleming worked in a London hospital where he studied ways to kill the bacteria that cause infections. For many of his experiments, he grew cultures of bacteria in petri dishes.

One morning, Fleming noticed an odd clear patch in the "lawn" of bacteria in one of the dishes. In a moment of carelessness, he must have accidentally knocked the cover off the dish and let some contaminant in. It was the biggest mistake of his life. Fortunately, he didn't ignore it.

Instead, thinking the clear patch might somehow be important, he examined it under a microscope. There, in the center of the clearing, he saw a tiny speck of mold. Since the area around the mold was bare of bacteria, Fleming hypothesized that the mold acted as a "poison pellet," oozing a substance that could kill the germs. This mold might be useful for treating bacteria infections, Fleming thought, but only if it proved to be harmless to people.

Experiment time. Fleming started by mixing the mold with human blood cells to see if it would kill them. It didn't. Then he tried injecting the mold into animals. Again, it was harmless. Later, when he used the mold to treat patients with infections, they were quickly cured.

Finally, he'd found an ideal germ killer, one that caused no harm to humans. Thanks to Fleming's flub, diseases like pneumonia, meningitis, and scarlet fever are no longer killers.