This is a fine account of the natural forces -- delicately balanced and constantly interacting -- that cause the vagaries of daily weather, and destructive events such as droughts, floods and extremes of heat and cold. It explains the growing recognition of the role in human affairs of climate and climatic variability, and concern about possible global climate change.

Gribbin, an astrophysicist by training, is the author of "Forecasts, Famines and Freezes" and a dozen other books of popular science. He shows that climatic patterns follow many cycles, some on geologic time scales, and that "normal" and "climate" must be defined: What area? What time period?

Gribbin cites studies ranging from paleobotany to volcanic dust veils, and gives considerable attention to sunspots -- though their link with climate is still elusive. His sketch of climatic history provides a necessary perspective. Variations in Earth's orbit around the sun can trigger ice ages, and he cites evidence that our present warm "inter-glacial" may be coming to an end: Over the next 10,000 years we may enter a new ice age.

His outline of the dynamics of natural change is useful in considering man-made "forcing" -- such as the steady buildup of carbon dioxide.

Earth's climate and atmosphere set it apart and make our planet a pleasant oasis in the cosmic void. Life is possible because we have liquid water, a blanket of life-supporting gases and a climate that is neither too warm nor too cold. Man may now be altering this equable climate by adding carbon dioxide to the atmosphere. This could make the global climate warmer than at any time in human history.

Carbon dioxide makes up only a tiny fraction of our atmosphere, but its importance far exceeds its relative scarcity. It is essential for photosynthesis, which converts the sun's energy into forms usable by plants and animals. It also helps regulate the vitally important heat balance. Water vapor and carbon dioxide act as a one-way screen, trapping part of the sun's heat and making the Earth's surface warmer than it otherwise would be. This is the so-called "greenhouse effect."

Our neighbor planets confirm the existence, nature and magnitude of the greenhouse effect, and the critical role of our atmosphere. On Mars, water is present only as ice, and its thin atmosphere allows most of the sun's energy to escape back into space. Mars is consequently a frozen desert. On Venus, however, the atmosphere is 96 percent carbon dioxide, and all its water is vaporized by the sun's heat. There, a "runaway" greenhouse pushes surface temperatures to nearly 500 C.

But man is changing the atmosphere, mainly by burning large quantities of oil, gas and coal, releasing carbon dioxide. We can only guess how much fossil fuel we may burn in the future. This will depend on many factors, including rates of population and economic growth, energy conservation and the development of nonfossil energy sources. We do know that carbon dioxide has increased by about 8 percent since careful measurement began in 1958, and by about 20 percent in a hundred years.

Mathematical models indicate that if these trends continue, we can expect an increase of about three degrees centigrade in average world temperature sometime in the next century. A few degrees doesn't sound like much compared with the daily and seasonal fluctuations, but a small change in average temperature can have a surprisingly large effect on global climate. On the human time scale, the change would seem irreversible.

The last Ice Age, for example, covered much of Europe and North America about 18,000 years ago, but average world temperature then was only five degrees centigrade cooler than at present.

A warmer Earth would mean a more active water cycle -- more evaporation and more rainfall. But these changes would not be evenly distributed: Some areas would become wetter; others, such as parts of the U.S. grain belt, might become drier. A carbon dioxide-induced climate shift could cause significant regional changes in world agricultural production.

In the past, man responded to climate change by moving to more suitable locations. Today, population growth and political boundaries make mass migration difficult. The climate may shift, but people may not be able to follow.

Modern societies are so finely tuned that any change could be disruptive, with possibly serious social consequences. Over time, man could adapt to a new climate regime. But the dislocations could be difficult and costly, even though the climate of some regions might improve. A global climate change could be the biggest environmental "impact" of all.

The carbon dioxide issue is full of troubling uncertainties. For one thing, natural climatic variability is so large we may have to wait a decade or two before the predicted "signal" of a carbon dioxide-warming emerges clearly from the background "noise." Even though the potential impacts are vast, we cannot precisely predict their timing, magnitude or location. But the uncertainties that make the question difficult do not make it less urgent.

Gribbin says there is no economical way to prevent the carbon dioxide buildup. Rather than searching for a "techno-fix," we should use the time and resources available to understand better the urgent problems of food and energy. Climate has always changed, he says, and it always will. Understanding past changes will help reduce our vulnerability to unfavorable climates of the future. He sees a possible carbon-dioxide greenhouse as another in an endless series of inevitable and normal fluctuations. The difference may be that this is a change we can, dimly at least, foresee. Perhaps we can prepare.