Cheaper than printing it out: buy the paperback book.

Inside this ultratight world -- hundreds of times more airtight than any NASA space capsule -- the atmosphere was full of surprises. It was unexpectedly clean for one thing. The trace gas buildup that was such a horrendous problem for earlier closed habitats and hi-tech closed systems such as NASA's space shuttle was eliminated by the collective respiration of a wilderness area. Scrubbed by some unknown balancing mechanism -- most probably microbial -- the air inside Bio2 was far cleaner than any space journey so far. Mark Nelson says, "Someone figured out it costs about $100 million a year to keep an astronaut in space, yet those guys are living in the worst environmental conditions you can imagine, worse than a ghetto." Mark told of an acquaintance who was honored to greet the returning space shuttle astronauts. She was nervously waiting in front of cameras as they readied the door. They opened the hatch. She got a whiff. She puked. Mark says, "These guys really are heroes, because they are living a lousy life."

For two years in Bio2, carbon dioxide levels meandered up and down. At one point during a six-day sunless period, CO2 reached a high of 3,800 parts per million (ppm). To give a sense of where that fits in, ambient carbon dioxide levels outside normally hover steadily at 350 ppm. The interior of a modern office building on a busy street may reach 2,000 ppm, and submarines let their CO2 concentration rise to 8,000 ppm before they turn on CO2 "scrubbers." Crew members of the NASA space shuttle work in a "normal" atmosphere of 5,000 ppm. Compare that to a very respectable 1,000 ppm average during a spring day in Bio2. The fluctuations, then, are well within the range of ordinary urban life and hardly noticeable to humans.

But the dance of atmospheric CO2 does have consequences on plants and the ocean. During the tense days of higher CO2, the biospherians worried that increased CO2 in their air would dissolve in the mild ocean water, increasing the formation of carbonic acid (CO2 + water), and lowering the water's pH, harming the newly transplanted corals. Discerning further biological effects of increased CO2 is part of the Biosphere 2 mission.

People pay attention to the makeup of the Earth's atmosphere because it seems to be changing. We are sure it is changing, but beyond that we know almost nothing about its behavior. The only measurement of any historical accuracy we have relates to one component: carbon dioxide. The information on CO2 concentration in the Earth's atmosphere shows an accelerating global rise over the past thirty years; that graph is due to a single, persistent scientist: Charles Keeling. In 1955, Keeling devised an instrument that could measure concentrations of carbon dioxide in all kinds of environments, from sooty city rooftops to pristine wilderness forests. Keeling obsessively measured CO2 anywhere he thought the level might vary. He measured CO2 at all times of the day and night. He initiated continuous measurements of CO2 on a Hawaiian mountaintop and in the Antarctic. A colleague of Keeling told a reporter, "Keeling's outstanding characteristic is that he has an overwhelming desire to measure carbon dioxide. He wants to measure it in his belly. Measure it in all its manifestations, atmospheric and oceanic. And he's done this all his life." Keeling is still measuring carbon dioxide around the world.

Keeling discovered very early that CO2 in the Earth's atmosphere cycles daily. CO2 in the air increases measurably at night when plants shut down photosynthesis for the day, and then hits a low in the sunny afternoon as the plants go full steam turning CO2 into vegetables. A few years later Keeling observed a second cycle: a hemispherical seasonal cycle of CO2, low in summer and peaking in the winter for the same reason CO2 peaks at night: no greens at work to eat it. But it is the third trend Keeling discovered that has focused attention on the dynamics of the atmosphere. Keeling noticed that the lowest level of CO2, no matter where or when, would never sink beyond 315 ppm. This threshold was the ambient, global CO2 level. And he noticed that every year it rose a little higher. By now, it's 350 ppm. Recently, other researchers have spotted in Keeling's meticulous recordings a fourth trend: the seasonal cycle is increasing in amplitude. It is as if the planet breathes yearly, summer (inhale) to winter (exhale), and its breath is getting deeper and deeper. Is Gaia hyperventilating or gasping?

Bio2 is a miniature Gaia. It is a small self-enclosed world with its own miniature atmosphere derived from living creatures. It is the first whole atmosphere/biosphere laboratory. And it has a chance to answer some of the tremendous questions science has about the workings of the Earth's atmosphere. Humans are inside the test tube to prevent the experiment from crashing, to divert the trials from overt crisis. The rest of us humans are outside, but inside the test tube of planet Earth. We are fiddling with Earth's atmosphere, yet haven't the slightest idea of how to control it, or where the dials are, or even if the system really is out of kilter and in crisis. The Bio2 experiment can offer clues to all those questions.

The atmosphere of Bio2 is so sensitive that the CO2 needle rises when a cloud passes over. The shade momentarily slows green manufacturing, which momentarily lets the input flow of CO2 back up, which immediately registers as a blip at the CO2 meter. On a partly cloudy day Bio2's CO2 graph shows a string of little atmospheric hiccups.

Despite all the attention CO2 levels have garnered in the past decade, and despite all the scrutiny agriculturists have given to the carbon cycle in plants, the fate of carbon in the Earth's atmosphere is a puzzle. It is generally agreed by climatologists that the curve of increasing CO2 within modern times very roughly matches the rates of carbon-burning by industrial humans. That neat fit leaves out one astounding factor: when measured more precisely, only half of the carbon now burned on Earth remains in the atmosphere as increased CO2 levels. The other half disappears!

Theories for the lost carbon abound. Three theories dominate: (1) it is being dissolved in the ocean, and then it precipitates to the sea bottom as carbon rain; or (2) it is being deposited in soils by microbes; or (3), most controversial, the lost carbon is fueling growth of the world's savanna grass, or being turned into tree wood, on an imperceptible but massive scale that we haven't yet been able to measure. CO2 is the accepted limiting resource for the biosphere. At 350 ppm, the concentration of carbon dioxide is only a faint .03 percent -- a mere trace gas. A field of corn in full sunshine will deplete the available trace CO2 within a zone three feet above ground in under five minutes. Even small increases in CO2 levels can boost biomass production significantly. Accordingly, says this hypothesis, wherever we aren't cutting down forests, trees are putting on extra weight due to the 15 percent of additional CO2 "fertilizer" in the air, perhaps even at a rate greater than they are being destroyed elsewhere.

So far, the evidence is confusing. But in April of 1992, two studies published in Science claimed that the ocean and biosphere of Earth are indeed stockpiling carbon at the scale needed. One article showed that European forests have gained 25 percent or more treeflesh since 1971 -- despite the negative effects of acid rain and other pollutants. But hardly anyone has looked at the global carbon budget in detail. Our global ignorance of the global atmosphere makes the Biosphere experiment very promising. Here in the relatively controlled conditions of a sealed bottle, the links between an operating atmosphere and a living biosphere can be explored and mapped.

The amounts of carbon in the atmosphere, in the soil, in the plants, and in the ocean of Bio2 were carefully measured before closure. As the sun heated up photosynthesis, the carbon was moved from air to living things by measurable amounts. Each time any plant material was harvested, it was laboriously weighed and recorded by the biospherians. They could perturb the system slightly to see how it changed. For instance, when Linda Leigh "turned on the savanna" with artificial summer rains, the biospherians made simultaneous measurements of carbon levels in all domains of subsoil, topsoil, air, and water. They compiled a rich chart of where all carbon lies at the end of two years. By saving dried samples of leaf clippings, they also traced (somewhat) the route that carbon traveled within the surrogate world by following shifts in the ratio of naturally occurring carbon isotopes.

Carbon was only the first mystery. But the riddle deepened. Oxygen levels were lower inside Bio2 than outside. Oxygen dropped from 21 percent of the Bio2 atmosphere to 15 percent. A 6 percent drop in oxygen concentration was equivalent to Bio2 being transported to a site at a higher elevation, with a thinner atmosphere. The residents of Lhasa, Tibet, thrive at a similar, slightly reduced oxygen level. The biospherians experienced headaches, sleep loss, and fatigue. Though not catastrophic, the drop in oxygen levels was bewildering. In a sealed bottle, where does disappearing oxygen go?

Unlike the lost-carbon riddle, the mysterious oxygen vanishing act in Bio2 was completely unexpected. Speculation was that oxygen in Bio2 was tied up in the newly minted soil, maybe being captured into carbonates formed by microorganisms. Or, perhaps the fresh concrete absorbed it. In a quick survey of the scientific literature, biospheric researchers found little data concerning atmospheric oxygen levels in the Earth's atmosphere. The only known (but little-reported) fact is that oxygen in the atmosphere of the Earth is most likely also disappearing! Nobody knows why or even by how much. "I am surprised that the general public all over the world is not clamoring to know how fast we are using up the oxygen," said visionary physicist Freeman Dyson, one of the few scientists to even raise the problem.

And why stop there? Several experts watching the Bio2 experiment have suggested that tracing the comings and goings of atmospheric nitrogen should be next. Although nitrogen is the bulk component of the atmosphere, its role in the Great Cycle is known only broadly. Like carbon and oxygen, what is known has been extrapolated from reductionist experiments in the lab and computer modeling. Others have proposed that the biospherians map the element sodium or phosphorus next. Generating big questions about Gaia and the atmosphere may be Bio2's most important contribution to science.

When the CO2 levels first began to rocket inside, the biospherians launched a countermove to limit the CO2 rise. The chief tool to leverage the atmosphere was deployment of an "intentional season." Take a dry, dormant savanna, desert or thorn scrub and rouse it into spring with rising temperatures. Soon a thousand leaf buds swell. Then pour on the rain. Bam! In four days the plants explode into leaf and flower. The awakened biome sucks up CO2. Once up, the biome can be kept awake past its normal retiring time by pruning old growth to stimulate new CO2-consuming growth. As Leigh wrote in late fall of the first year, "With short days of winter approaching, we have to prepare for reduced light. Today we began to prune back the ginger belt on the north edge of the rain forest in order to stimulate rapid growth -- a routine atmosphere management task."

The humans managed the atmosphere by turning the "CO2 valve." Sometimes they reversed it. To flood the air with carbon dioxide, the biospherians hauled back the tons of dried grass clippings they had removed earlier. The clippings were piled on the soil as mulch and wetted. As bacteria decomposed it, they released CO2 into the air.

Leigh called the biospherian interference in the atmosphere a "molecule economy." When they coordinated the atmosphere, they would "deposit the carbon into our account for safekeeping so that we can spend it next summer when we will need it for long days of plant growth." The underground areas where the plant clippings were dried served as a carbon bank. Carbon was lent as needed and primed with water. Water in Bio2 was diverted from one locality to another like so much federal spending meant to stimulate a regional economy. By channeling water onto the desert, CO2 shrank; by channeling the water onto the dried mulch, CO2 expanded. On Earth, our carbon bank is the black oil under Arabian sands, but all we do is spend it.

Bio2 compressed geological time into years. The biospherians twiddled with "geological" adjustments of carbon-storing and withdrawing carbon atoms in bulk -- in the hope of roughly tuning the atmosphere. They tinkered with the ocean, lowering its temperature, adjusting the return of salty leachate, nudging its pH, and simultaneously guessing on a thousand other variables. "It's those few thousand other variables that make the Bio2 system challenging and controversial," said Leigh. "Most of us are taught not to mess with even two simultaneous variables." The biospherians hoped that if they were lucky, they could temper the initial wild oscillations of the atmosphere and ocean in the first years with a few well-chosen drastic actions. They would be the training wheels until the system could cycle through the year relying only on the natural action of sun, seasons, plants and animals to keep it in balance. At that point the system would "pop."

"Pop" is the term hobbyists in the saltwater aquarium trade use to describe what happens when a new fish tank suddenly balances after a long, meandering period of instability. Like Bio2, a saltwater fish tank is a delicate closed system that relies on an invisible world of microorganisms to process the waste of larger animals and plants. As Gomez, Folsome, and Pimm discovered in their microcosms, it can take 60 days for the microbes to settle into a stable community. In aquariums it takes several months for the various bacteria to develop a food web and to establish themselves in the gravel of the start-up tank. As more species of life are slowly added to the embryonic aquarium, the water becomes extremely sensitive to vicious cycles. If one ingredient drifts out of line (say, the amount of ammonia), it can kill off a few organisms, which decompose to release even more ammonia, killing more creatures, thus rapidly triggering the crash of the whole community. To ease the tank through this period of acute imbalance, the aquarist nudges the system gently with judicious changes of water, select chemical additives, filtration devices, and inoculations of bacteria from other successful aquariums. Then after about six weeks of microbial give-and-take-the nascent community teetering on the edge of chaos -- suddenly, overnight, the system "pops" to zero ammonia. It's now ready for the long haul. Once the system has popped, it is more self-sustaining, self-stabilizing, not requiring the artificial crutches that set-up needed.

What is interesting about a closed-system pop is that the conditions the day before the pop and the day after the pop hardly change. Beyond a little babysitting, there is often nothing one can do except wait. Wait for the thing to mature, to ripen, to grow, and develop. "Don't rush it," is the advice from saltwater hobbyists. "Don't hurry gestation as the system self-organizes. The most important thing you can give it is time."

Still green after two years, Bio2 is ripening. It suffers from wild, infantile oscillations that require "artificial" nurturing to soothe. It has not popped yet. It may be years (decades?) before it does, if it ever does, if it even can. That is the experiment.

We have not really looked yet, but we may find that all complex coevolutionary systems need to pop. Ecosystem restorationists such as Packard on the prairie and Wingate on Nonsuch Island seem to find that large systems can be assembled by ratchetting up complexity; once a system reaches a level of stability it tends not to easily fall back again, as if the system was "attracted" by the cohesion birthed by the new complexity. Human institutions, such as teams and companies, exhibit pop. Some little nudge -- the additional right manager, a nifty new tool -- can suddenly turn 35 competent hard-working people into a creative organism in the state of runaway success. Machines and machine systems, once we build them with sufficient complexity and flexibility, will also pop.