The theorized companion star, through its gravitational pull, unleashes a furious storm of comets in the inner solar system lasting from 100,000 to 2 million years.
Several of these comets strike the Earth.
"Heavy snows are driven and fall from the world's four corners; the murder frost prevails. The Sun is darkened at noon; it sheds no gladness; devouring tempests bellow and never end. In vain do men await the coming of summer. Thrice winter follows winter over a world which is snow-smitten, frost-fettered, and chained in ice.

"Fimbul Winter" from Norse saga, Twilight of the Gods

Our species, Homo sapiens, arose approximately 250,000 years ago. In the beginning, we used tools of stone and sought shelter in caves. Today, our shelters scrape clouds and our tools allow us to see galaxies far beyond our own, or peer deep into the heart of matter itself. So much progress in such a short time, for in geological terms, the reign of our species has been but the proverbial blink of an eye. Imagine, however, what our record of achievement would be had our history been disrupted no less than five times by titanic nuclear wars, each delivering a destructive blast 10,000 times more powerful than the combined yield of all existing nuclear weapons in our world today.

Such upheaval is what many other species, including the dinosaurs, may have faced during the history of our planet, according to a theory set forth by a Lawrence Berkeley Laboratory (LBL) scientist and his colleagues. The theory postulates that every 26 to 30 million years, life on Earth is severely jeopardized by the arrival of a small companion star to the sun. Dubbed "Nemesis" (after the Greek goddess of retribution), the companion star through its gravitational pull unleashes a furious storm of comets into the inner solar system that lasts anywhere from 100,000 years to two million years. Of the billions of comets sent swarming toward the sun, several strike the Earth, triggering a nightmarish sequence of ecological catastrophes.

"We expect that in a typical comet storm, there would be perhaps 10 impacts spread out over two million years, with intervals averaging 50,000 years between impacts," says LBL astrophysicist Richard Muller. In 1984, Muller, along with UC Berkeley astronomer Marc Davis and Piet Hut, an astronomer with the Institute for Advanced Study at Princeton University, announced the Nemesis theory in Nature magazine. As could be expected, it was and remains controversial. However, although the evidence for the existence of Nemesis is still circumstantial, this evidence continues to mount, and the theory has so far withstood all challenges.

Nemesis was the culmination of a chain of events that began in 1977, in Gubbio, Italy, a tiny village halfway between Rome and Florence. Walter Alvarez, a UC Berkeley geologist, was collecting samples of the limestone rock there for a study on paleomagnetism. The limestone rock outside of Gubbio is a big attraction for geologists and paleontologists because it provides a complete geological record of the end of the Cretaceous period and the beginning of the Tertiary period. This transition took place 65 million years ago, and is of special significance to our species, for it marked the close of the "Age of Reptiles," when dinosaurs ruled the Earth. Sometimes referred to as "the Great Dying," the massive extinction that engulfed the dinosaurs claimed nearly 75 percent of all the species of life on our planet, including most types of plants and many types of microscopic organisms. As much as 95 percent of all living creatures might have perished at the peak of destruction.

Sandwiched between the limestone of the two periods, forming a clear line of demarcation, is a thin maybe one-half-inch thick layer of red clay. Immediately below this clay layer, the Cretaceous limestone is heavily populated with a wide mix of the tiny fossils of marine creatures called forams. Above the clay layer, in the Tertiary limestone, however, the fossils of but a single species of foram can be seen. The clay layer itself contains no foram fossils at all.

When Walter Alvarez brought his samples back to Berkeley, his father, LBL Nobel laureate physicist Luis Alvarez, suggested that subjecting them to neutron activation analysis could help determine how long it took for the clay layer to form. The analysis, performed by LBL nuclear chemists Frank Asaro and Helen Michel, revealed to the surprise of everyone involved that the clay was about 600 times richer in iridium than the surrounding limestone. A silvery-white metal, related to platinum, iridium is quite scarce in the Earth's crust, found usually in concentrations of only 20 parts per trillion. When the Earth was formed, most of the iridium sank into the planet's core, 3,000 miles below the surface, where the concentration of the metal is 10,000 times that in the crust. Other sources of high iridium concentrations are extraterrestrial objects, such as meteorites or comets.

Subsequent samples collected from clay layers found at locations in Denmark and New Zealand, where the geological record of the Cretaceous-Tertiary boundaries are also complete, revealed the same iridium anomaly, plus an abundance of soot. This iridium anomaly has now been identified at more than 75 sites worldwide, by scientists from 11 different laboratories. Iridium is generally found in combination with platinum, gold, and several other elements. Measuring the concentrations of these elements and comparing their ratio to iridium indicated that the widely scattered iridium all came from the same source.

Putting all of the data together, Luis Alvarez concluded that the iridium anomaly was the result of a collision between the Earth and an extraterrestrial object approximately six miles in diameter. He speculated further that it was this collision that led to the death of the dinosaurs and all of the other species that perished during the Great Dying.

When a rock the size of San Francisco, traveling at approximately 45,000 miles per hour, hits the Earth, there is an instantaneous release of approximately 100 million megatons of kinetic energy six billion times the force of the Hiroshima bomb. Luis and Walter Alvarez predicted the effects of such an explosion, based on the aftermath of the volcanic eruption of Krakatoa in 1883, the biggest eruption ever recorded.

If the impact takes place on land, a heavy shroud of fine dust particles from the shattered planetary crust and the pulverized meteorite or comet would be swept high into the stratosphere by the mushrooming fireball, where it would slowly spread, wrapping the entire globe in a dense cocoon. The fireball's blazing heat would ignite enormous wildfires, the soot and debris from which would rise up and add to the sky-blackening dust, creating an extended period of endless night.

Said Walter Alvarez in a report for the American Geophysical Union, "For a few months, it would be so dark you literally could not see your hand in front of your face."

The darkness would shut down the photosynthetic process, killing all but the hardiest of plant species and driving the food chain into a state of collapse. Worldwide starvation would ensue as animals that feed on the plants die and the predators in turn follow. Extremely cold temperatures brought on by the darkness might usher in an ice age. Even if the impact takes place in the ocean, dust (from the crushed ocean floor) would still be shot above the atmosphere, only accompanying the dust would be tremendous volumes of vaporized water. After the dust finally settled, the water vapor would still remain. Solar heat reflected off the Earth's surface would be prevented from escaping into outer space by this thick moisture, and the consequence would be an oppressive greenhouse effect.

"The bitter cold would be followed by a sweltering heat," said Walter Alvarez in his AGU report.

To make matters worse, the energy released by the impact could serve as a catalyst to combine atmospheric nitrogen and oxygen into nitric acid that would fall back on the surface as corrosive precipitation.

Singular event or an event that has recurred

A plot of data on life extinctions, collected by David Raup and John Sepkoski at the University of Chicago, shows peaks in the extinction rate occurring at 26 to 30 million year intervals, as indicated by arrows
As originally proposed, the Alvarezes saw the Great Dying and the iridium anomaly as a singular event a fluke in Earth's history.

A second iridium anomaly was discovered in samples taken from sediment that had been deposited on the floors of the Caribbean Sea and the Gulf of Mexico about 35 million years ago, when a less severe extinction occurred, but no one proposed a link between the two events. Then, in 1984, came a report from two University of Chicago paleontologists, David Raup and John Sepkoski, who had put together a detailed list of sea life that had become extinct during the past 250 million years. Containing more than 3,500 different species, it was the most complete extinction list ever compiled. When they subjected their list to computer analysis, Raup and Sepkoski discovered that mass extinctions occur periodically, approximately every 26 to 30 million years.

Scientists immediately scrambled to find an explanation that could account for a persistent, recurring cycle of planet-wide species die-outs. Volcanic eruptions were the most obvious suspects, but volcanoes fail to account for the clay layer, the high soot content and, most significantly, the high iridium concentrations. Casting further doubt on the culpability of volcanoes was the discovery of shock quartz and microtektites along with the iridium and soot in the clay layer samples taken from around the world.

Shock quartz silt-sized grains of quartz, which, under a microscope, show cracks and strains, is formed in the heat and pressure of a powerful explosion. It showed up routinely in rocks brought back from the moon by the lunar astronauts, but on Earth it has been found only in meteorite craters and at nuclear weapon test sites. Microtektites are tiny pieces of glass, believed to be droplets of rock that were melted in the heat of an impact and hurled up beyond the atmosphere where they cooled. Upon reentry, the droplets were reheated. The heating-cooling-reheating sequence gave the microtektites in the clay layer a unique spherule shape. Violent volcanic eruptions, such as took place on Mt. St. Helens, Washington, in 1980, can produce glassy material, but always in angular shapes because the melted rock is never ejected beyond the atmosphere. The quiet eruptions of the gentle basaltic volcanoes, prominent in Hawaii, will cough up spherule-shaped glass, called "Pele's tears," but distribute the material only in the immediate vicinity.

Ruling out other terrestrial causes, many scientists turned to the heavens. One possibility was meteorites, which are chips of asteroids or planets moving randomly through space. However, a mechanism to explain the periodicity of the extinctions has yet to be found. A second possibility was comets, "dirty snowballs" of ice with a rocky center. Looping the solar system, beyond the orbit of Pluto and extending out to more than eight trillion miles, is a vast bracelet of comets known as the "Oort cloud," after its discoverer, Dutch astronomer Jan Oort. The trillions of comets in the Oort cloud generally maintain a slow but steady orbit around the sun. Occasionally, the gravitational field of a passing star will jar some comets loose, but few of these ever reach the inner solar system (Mercury, Venus, Earth, and Mars), as the gravitational pulls of Jupiter and Saturn acting somewhat like giant vacuum cleaners keep this part of the system relatively clean of comets and other space debris.

However, a strong enough gravitational force could dislodge so many comets that, through sheer numbers, the inner solar system's protective cleaning mechanism would be overwhelmed. One of the first possible sources of this gravitational force to be considered was the molecular dust clouds in the central plane of the Milky Way. As the solar system revolves around the center of the galaxy, it bobs up and down, periodically crossing through the star-crowded central plane that is foggy with molecular dust star stuff that never coalesced. One of the many problems with this suggestion is that measurements have shown the molecular dust clouds to be far too thinly dispersed to exert sufficient tidal gravitational force. Also, the bobbing of the sun does not match the times of extinction, in fact, the sun is close to the central plane right now.

Another source of gravitational pull that has been proposed is the existence of a tenth planet in the solar system. Called "Planet X," this planet would be a gas ball as much as five times the size of Earth, occupying a peculiar shifting orbit that is tilted at an angle to the solar plane of the nine known planets. This theory also calls for the existence of an as yet undetected inner disk of the Oort cloud, between the orbits of Neptune and Pluto. Every 26 to 30 million years, the orbit of Planet X would be shifted so that it would scrape the edge of the inner disk, sending a host of comets towards the sun. The major problem with this proposal is that the hypothetical inner disk of the Oort cloud would be unstable and could not remain a disk. Consequently, comets would be shaken loose in a steady shower over the 26 to 30 million year time periods, rather than torn loose in a concentrated storm.

A mechanism to explain the periodicity of the extinctions

The Nemesis theory fulfills all the requirements prescribed by the Raup and Sepkoski mass extinction timetable.

As envisioned by Muller, Davis, and Hut, Nemesis is probably a red dwarf, the most common type of star in the galaxy (three-fourths of all the stars in the Milky Way are believed to be red dwarfs). Less than a third the size of the sun and about one one-thousandth as bright, Nemesis might travel in an elliptical orbit that at its perihelion (closest point) brings it within a half light year of the sun (one light year is about six trillion miles) and into the midst of the Oort Cloud. Right now, Nemesis may be at its aphelion (most distant point), nearly three light years away. The sun's closest known neighbor, Proxima Centauri, is about 4.25 light years distant.

Another group of scientists, led by Daniel Whitmire, an astrophysicist with the University of Southwestern Louisiana, and Al Jackson, of the Computer Science Corporation, announced their own theory of a companion star to the sun in the same issue of Nature as Muller and his colleagues. Although the means of triggering massive extinctions are essentially the same, this second group believes the companion star is invisible: either a brown dwarf, a star so tiny that it never ignited, or a black hole, a shrunken star so dense that its gravity prevents any light from escaping.

"We see no reason to assume the star is invisible," says Muller, "since most of the stars in the sky have never had their distance from us measured. If the companion has a magnitude between 8 and 12, it would be dim enough to have been missed in full sky parallax surveys."

That the sun would have a companion star is by no means farfetched. More than 50 percent of the stars in the galaxy are partners in a binary system. The elliptical orbit of Nemesis would carry it farther away from the sun than the distance separating companions in any known binary system. Some scientists have protested that this orbit is too elliptical to be maintained and that Nemesis would have long since left the system. However, the calculations of Hut show Nemesis' orbit being stable for about a billion years.

Says Muller, "The stability of the orbit is sufficiently long to account for the regularity in the extinctions, but it also implies that the companion star could not have been in this orbit since the formation of the Earth. Since gravitational capture is very improbable, the most likely scenario is that the companion star was once more tightly bound to the sun and its orbit is slowly being dissipated by passing stars."

It is even possible, Muller suggests, that the gravitational shoving of Nemesis out into a more distant orbit coincided with an event referred to by astronomers as "the late great bombardment." Approximately four billion years ago, a celestial version of saturation bombing left the surface of the moon badly scarred with craters, which, because of the absence of atmospheric erosion, can still be seen. Voyager has shown the moons of Mars, Jupiter, and Saturn to be similarly pocked.

The first fossil records on Earth also date back four billion years ago. Mysteriously enough, the division between the Earth's two eons, the Cryptozoic eon ("hidden life") and the Phanerozoic eon ("visible life") is sharply etched. Rather than a gradual appearance of increasingly complex fossils, the records show that the Cryptozoic eon ends with no fossils at all above the microscopic level, then the Phanerozoic eon begins and suddenly a dozen different types of elaborate organisms materialize.

Testing the theory

When Muller told Walter Alvarez about the Nemesis theory, the younger Alvarez saw that one means of testing it would be an examination of impact craters on Earth. If the theory is correct, craters should be clumped together in periodic segments of time corresponding to the times that mass extinctions took place. Unlike on the airless moon, where craters are preserved in near perpetuity, on the Earth, most craters are erased by water and wind erosion, as well as continental drift. However, some do survive, about a hundred of which are known. Examining 13 of the largest, most accurately dated of these craters, spanning the 250 million years of the mass extinctions studied by Raup and Sepkoski, Muller and Alvarez found the same 26 to 30 million year periodicity.

"Our analysis has proven to be rather robust against changes in the data set," says Muller, "including the addition or elimination of a few craters, or changes in the minimum crater diameter examined."

Recently, Muller and LBL physicist Saul Perlmutter used cosmic ray exposure ages to show that meteorites created by comets fell on Earth in flurries at approximately the dates of the last three major extinctions.

"Exposure to cosmic rays begins when a meteorite is broken out of the parent body that had previously shielded it, usually an asteroid or the planet Mars, and ends when the meteorite lands on Earth," says Muller. "The cosmic ray exposure age of a meteorite can be determined by the amount of certain isotopes, such as neon 21, which are produced at a known rate by energetic cosmic rays hitting the meteorite. This exposure age tells us the time the meteor spent orbiting in the solar system since its creation."

There are two types of meteorites, high-iron and low-iron. The high-iron meteorites (28 percent by weight), called "H chondrites," are formed when material from the iron-rich core of an asteroid or planet is blown out into space by a violent collision with a speeding comet. Low-iron meteorites, or "L chondrites," are formed from surface material tossed out by low-velocity collisions between asteroids. During their periodic flurries, high-iron meteorites fall on Earth in much greater numbers than low-iron meteorites, but in between these periods, the number of high- and low-iron meteorites striking Earth is about the same.

"The distribution of the H chondrite cosmic ray ages provides new evidence confirming the claim of comet storm theory that a large fraction of the impacts on the Earth occur during relatively brief periods," says Muller. "This is the first evidence for comet storms not based on terrestrial effects."

The evidence for Nemesis-triggered periodic comet storms based on cosmic ray exposure ages was drawn primarily from reviews of existing data. "In these days of tight budgets," observes Muller wryly, "the cheapest way to do research is in the library." Another review of existing data, this time by Muller and LBL physicist Donald Morris, uncovered evidence for periodic comet storms in the Earth's magnetic field.

Volcanic rock, as it cools from the lava state, aligns itself with the Earth's magnetic field. In 1906, French physicist Bernard Brunhes discovered volcanic rock magnetized in the opposite direction of today's field. It is now known that the Earth's magnetic field has reversed itself many times throughout the planet's history, and at times has even been switched off. Muller and Morris felt that at least some of these geomagnetic flips were caused by comet impacts, and they developed a model to explain how it happened.

The Earth's magnetic field is generated by slow eddies in its molten nickel-iron core that are the product of the heat flow out of the core, modified by the planet's rotation. When a crashing comet plunges the world into darkness, temperatures on the land drop much faster than those in the sea because water retains heat much longer than soil. According to the model of Muller and Morris, water near the equator evaporates and is redistributed as ice and snow on the polar caps. The result is a sudden (within a few hundred years) drop in the level of the oceans. In accordance with the conservation of angular momentum, the redistribution of mass alters the rotation rate of the Earth's crust and mantle with respect to the liquid core and leads to a disruption of the magnetic field.

"It is the same as when figure skaters go into a spin with their arms extended, then draw their arms in to increase their rotational speed," says Muller. "The Earth's magnetism is so sensitive to the motions of the liquid core that it doesn't take much of a change in rotational rate to affect the field."

Prior to the work of Muller and Morris, Chicago's Raup had examined the frequency of 296 geomagnetic reversals that took place during the last 170 million years and found peaks in the rate of reversals occurring approximately every 30 million years. Deposits of microtektites were also found in volcanic and seabed rocks from times when reversals took place. There was a sudden drop in sea level during the die-out of the dinosaurs, but there is no evidence of a geomagnetic reversal. This does not blemish the model of Muller and Morris, however, for it predicts that magnetic excursions, during which the field is turned off, would result from half of the impacts. Magnetic excursions are difficult to detect in volcanic rock.

"Our model readily explains observed geophysical correlations, and accounts for the behavior of the Earth's magnetic field during a reversal," says Morris. "Although somewhat speculative, it is based on assumptions that are considered plausible by experts in the relevant scientific fields."

A geomagnetic reversal could also take place should the polar caps melt and cause a sudden swelling of the seas. This, too, would alter the rotation of the Earth's crust and mantle with respect to the core and disrupt the dynamo.

The Nemesis scenario

When Luis and Walter Alvarez first presented their idea that the impact of an extraterrestrial object sparked the death of the dinosaurs, many paleontologists were quick to protest that the extinction of the dinosaurs did not transpire within a year or two, but was a gradual decline that dragged on for several hundred thousand years. Nemesis-launched comet storms reconcile this apparent contradiction.

"We would not necessarily expect all species to die out simultaneously during a storm," says Muller. "Some species would be destroyed by an early impact, while others make it through, only to be killed by a later and larger impact."

Under the Nemesis scenario, what at first glance might appear to be a single, gradual extinction, would, upon closer scrutiny, turn out to be a series of individual, abrupt, mass die-outs. This picture fits closely with the new school of evolutionary thought, coined "punctuated equilibrium" by Harvard paleontologists Steven Jay Gould and Niles Eldredge. In contrast to Charles Darwin's view of evolution being a steady process of smooth transitions to ever higher forms of life, what fossil records actually show are long stretches of inactivity, then a sudden jump over a few hundred generations.

"It is as if evolution has its own kind of death, giving new species a chance," says Muller. "The species-versus-species competition that Darwin proclaimed appears to take place only during the relatively quiet periods between comet storms. Every 26 million years or so, instead of survival of the fittest, we may be looking at survival of the first, where the species that fills an open ecological niche first has the advantage. Without this catastrophe mechanism, Earth might still be a world ruled by trilobites."

The extinction of the dinosaurs is the best illustration of the effect a Nemesis companion star could have on our planet's history. For years, school children were taught that the dinosaurs died out because they were cold-blooded clods, too big, too bulky, too slow, and too stupid to adapt to changing environmental conditions and competition from swift, small, clever, egg-eating mammals. This orthodoxy conveniently overlooked the fact that dinosaurs co-existed and ruled over mammals for more than 100 million years, 400 times longer than the reign of Homo sapiens. At the height of their glory, during the Cretaceous period, the menagerie of different dinosaurs filled nearly every ecological niche. When they were toppled, the ecological slate was wiped clean and mammals rapidly diversified to refill it.

"Why are we here?" Steven Jay Gould has asked. "Because the dinosaurs disappeared, not because the mammals out-competed them."

Search for Nemesis

For now, Nemesis is a tantalizing specter. The case for the companion star is perhaps solid enough to score a victory in a court of law, but in the court of science, the ultimate proof will be in the finding. Joining Muller in the search for Nemesis at LBL are Perlmutter and physicists Carl Pennypacker, Frank Crawford, and Roger Williams. Using the computer-controlled 30-inch reflecting telescope at Leuschner Observatory, in Lafayette, Calif., the scientists are in the process of photographing 5,000 red stars in the northern hemisphere and measuring the parallax of each the shift in its apparent position as the Earth rotates around the sun. The telescope has been programmed to photograph each candidate, wait two to six months, then photograph each star a second time. The two positions can then be compared. A star far away will show little if any change in position, but a star close enough to be Nemesis will have moved perceptibly.

So far, the Nemesis search has eliminated 41 stars. Says Perlmutter, "The system was difficult to start, but we've got it down now and could soon have the data on 3,000 more stars." It is Muller's suspicion that Nemesis might well be hiding in a constellation in the southern hemisphere called Hydra, simply because," he muses, "It's the biggest."

Terrestrial-based testing of the Nemesis theory also continues. The presence of an iridium anomaly in craters that correspond to mass extinctions, and in volcanic rocks and sea beds that correspond to geomagnetic reversals would be a strong supporting argument for the occurrence of comet storms. Sediment samples are now being collected from far-flung locales and send to LBL's Asaro and Michel for analysis. The analysis process should go much faster than ever before with the use of a new detection device called the "Iridium Coincidence Spectrometer." Conceived by Luis Alvarez and designed by Asaro, the ICS should do in three years what previous equipment would have taken more than 100 years to do. Asaro and Michel expect to be able to analyze 6,000 samples a year.

Humanity has never had to face the megablast of even one major comet impact. Perhaps the most important aspect of the Nemesis theory, and the one for which we as a species can be most grateful, is that the deadly little companion star is not due to return until the year 15 million A.D.

Additional Information: