The history of science contains more than a few instances in which an old idea or hypothesis, considered improbable, dubious or merely wrong by modern investigators, is, through a combination of new evidence and a more thorough look at the old evidence, shown to be correct after all. Something like this seems to be happening to modern science's view of the sun. The current dogma is that the sun is steady, dependable, constant. In this view, its well-known 11-year sunspot cycle is the manifestation of a smoothly running, well-ordered machine, clicking with regularity like astrophysical clockwork. It is a comfortable view, the sun being of some importance to us all here on earth. Now an astronomer with a historical bent has delved back through past observational records and, by making numerous independent cross checks, resurrected and made a persuasive case for an old hypothesis that the solar cycle and the sun itself have changed in historic time. The evidence shows that for a 70-year period from A.D. 1645 to 1715 sunspots were almost totally absent on the surface of the sun. Solar activity was at a near-zero level, a true and strange anomaly. "This is totally unlike the modern behavior of the sun," says the astronomer, John A. Eddy of the National Center for Atmospheric Research's High Altitude Observatory in Boulder, Colo., "and the consequences for solar and terrestrial physics seem to me profound." The evidence shows not only a minimum in solar activity from A.D. 1645 to 1715 but also an earlier minimum from A.D. 1460 to 1550 and an even earlier maximum from A.D. 1 100 to 1250. "We now have to realize," Eddy says, "that the sun's behavior has been better in the last 200 years than in the previous 1,000 years. " The view that solar activity can and has varied to such a major degree in long-term patterns in historic time alters widely accepted assumptions about the constancy of the sun. "We've shattered the Principle of Uniformitarianism for the sun," Eddy says. By this he means that the present behavior of the sun can no longer be considered a reliable guide to the behavior of the sun in the past. Eddy's conclusions imply that the often-discussed 11-year solar cycle is of far less importance and concern than are longer term variations-the overall "envelope" of solar activity. That patterns of solar activity have varied over historic time is interesting enough in itself. But beyond that, Eddy believes that the longterm fluctuations may be due to changes in the solar constant, the total radiative output of the sun. Such an idea is of fundamental importance. Whether the solar constant may vary, once considered improbable, is now being much debated. The problem has taken on new significance as solar physicists and climatologists consider the possible effects of the sun on variations in earth's climate. All this becomes even more intriguing when one observes that the period of near total absence of solar activity from 1645 to 1715 coincides almost precisely with the coldest point in the climatic minimum on earth that we now call the Little Ice Age. As Eddy puts it, "The climate curve looks a lot like the curve of variability in solar activity." Eddy described his results in a session on the sun's effects on terrestrial climate at the annual meeting of the American Association for the Advancement of Science in Boston last week (his paper will appear soon in SCIENCE). When he finished, moderator George B. Field, director of the Center for Astrophysics of the Smithsonian Astrophysical Observatory and the Harvard College Observatory in Cambridge, turned to the small but crowded lecture room and said, "Maybe we've heard a turning point in the history of science." The main contribution of Eddy's analysis is to show that the prolonged sunspot minimum beginning in the 17th century is not an artifact of incomplete or spurious data but is in fact real. This has been the major stumbling block to acceptance of the idea, which has been around since at least the late 19th century. It was then that two well-known solar astronomers, Gustav Sporer of Germany in papers published in 1887 and 1889 and E. W. Maunder of the Greenwich Observatory in more detailed papers published in 1890 and 1894, called attention to the 70-year absence of sunspots. Maunder emphasized