Walter Gratzer's "The Undergrowth of Science" is about those times when scientific claims do not hold up. Scientists are quick to point out that this is uncommon, that science is a self-correcting process in which truth and reality are determined by the repeatability of hands-on experiments and the predictive power of abstract theories. What they don't say is that science needs to be self-correcting because of scientists themselves. Like people in other professions, scientists can be susceptible to self-deception, conceit, wishful thinking, ideological coercion and fraud.

Scientific fraud is an interesting phenomenon in itself. Books have been written about several famous cases, and more will surely follow. In October, for example, a Japanese archaeologist admitted to planting artifacts at a dig site so he could claim to have discovered the oldest tools in Japan. No Indiana Jones, he was caught on videotape. But Gratzer is interested in the unintentional deceptions that creep into science, and the political and psychological necessities that compel some scientists to cling to false beliefs in the face of common sense and scientific evidence to the contrary.

Scientists first have to mislead themselves before they can mislead others. Nationalism, or "tribal delusion," often helps both levels of deception, as Gratzer shows in the famous case of the German rays vs. the French rays. At the end of the 19th century, scientists were beginning to dissect the mysterious worlds of radioactivity, light and the atom. In 1895, a German physicist named Wilhelm Roentgen discovered a powerful new radiation he called "X-rays." Rene Blondlot, a highly regarded French physicist studying Roentgen's X-rays, convinced himself he had discovered another form of radiation. He named this radiation "N-rays" after his university in Nancy. Then he used his experiments to convince other French scientists. Foreign scientists, including the Germans, were skeptical. Partly because of German criticism, the French quickly rallied around Blondlot. The Germans had invaded Paris 30 years earlier, French lands had been annexed and both sides were still fighting pitched academic battles of "Our science is better than your science." The cross-border war of rays ended in 1904 when an American physicist who was unable to repeat Blondlot's observations pointed out his experimental errors and subjective interpretations of the results. Blondlot, however, quietly continued to work on his imaginary N-rays until his death in 1930.

Another fictitious entity that occupied many competent scientists was polywater. A Russian chemist discovered this viscous form of ordinary water in the 1960s. Its unusual properties worried some people (at least those who had read Kurt Vonnegut's tale about a strange form of ice, ice IX, that froze the world's oceans). Could polywater turn the oceans into a viscous jello? Would it impede Navy ships and end surfing? Hundreds of scientists began studying polywater and publishing papers on it. But research came to a screeching halt when Bell Labs' Dennis Rousseau wrung his sweaty gym shirt into a glass tube and discovered that his salty, protein-laced sweat had the same properties as polywater.

Maya Angelou reads from "The Heart of a Woman"