Unofficial SJG Archive

The Unofficial Stephen Jay Gould Archive

Unofficial SJG Archive

Planet of the Bacteria

by Stephen Jay Gould

y interest in paleontology began in a childhood fascination with dinosaurs. I spent a substantial part of my youth reading the modest literature then available for children on the history of life. I well remember the invariant scheme used to divide the fossil record into a series of "ages" representing the progress that supposedly marked the march of evolution: the "Age of Invertebrates," followed by the Age of Fishes, Reptiles, Mammals and, finally, with all the parochiality of the engendered language then current, the "Age of Man."

I have watched various reforms in this system during the past 40 years. The language police, of course, would never allow an Age of Man any more, so we could, at best and with more inclusive generosity, now specify an "age of humans" or an "age of self-consciousness." But we have also come to recognize, with even further inclusive generosity, that one species of mammals, despite our unbounded success, cannot speak adequately for the whole. Some enlightened folks have even recognized that an "age of mammals" doesn't specify sufficient equity—especially since mammals form a small group of some 4,000 species, while nearly 1 million species of multicellular animals have been formally named. Since more than 80 percent of these million are arthropods and since the great majority of arthropods are insects, these same enlightened people tend to label modern times as the "age of arthropods."

Fair enough, if we wish to honor multicellular creatures, but we are still not free of the parochialism of our scale. If we must characterize a whole by a representative part, we certainly should honor life's constant mode. We live now in the "Age of Bacteria." Our planet has always been in the "Age of Bacteria," ever since the first fossils—bacteria, of course—were entombed in rocks more than 3 billion years ago.

On any possible, reasonable or fair criterion, bacteria are—and always have been—the dominant forms of life on Earth. Our failure to grasp this most evident of biological facts arises in part from the blindness of our arrogance but also, in large measure, as an effect of scale. We are so accustomed to viewing phenomena of our scale—sizes measured in feet and ages in decades—as typical of nature.

Individual bacteria lie beneath our vision and may live no longer than the time I take to eat lunch or my grandfather spent with his evening cigar. But then, who knows? To a bacterium, human bodies might appear as widely dispersed, effectively eternal (or at least geological), massive mountains, fit for all forms of exploitation and fraught with little danger unless a bolus of imported penicillin strikes at some of the nasty brethren. Consider just some of the criteria for bacterial domination:


The fossil record of life begins with bacteria, at least 3.5 to 3.6 billion years ago. About half the history of life later, the more elaborate eukaryotic cell makes a first appearance in the fossil record—about 1.8 to 1.9 billion years ago by best current evidence.

The first multicellular creatures—marine algae—enter the stage soon afterward, but these organisms bear no genealogical relationship to our primary interest: the history of animal life. The first multicellular animals do not enter the fossil record until about 580 million years ago—after about five-sixths of life's history had already passed. Bacteria have been the stayers and keepers of life's history.


Let us make a quick bow to the flip side of such long domination to the future prospects that match such a distinguished and persistent past. Bacteria have occupied life's mode from the very beginning, and I cannot imagine a change of status, even under any conceivable new regime that human ingenuity might someday impose upon our planet.

Bacteria exist in such overwhelming number and such unparalleled variety; they live in such a wide range of environments and work in so many unmatched modes of metabolism. Our shenanigans, nuclear and otherwise, might easily lead to our own destruction in the foreseeable future. We might take most of the large terrestrial vertebrates with us—a few thousand species at most.

I doubt that we could ever substantially touch bacterial diversity. The modal organisms cannot be nuked into oblivion or very much affected by any of our considerable conceivable malfeasances.


The history of classification for the basic groups of life is one long tale of decreasing parochialism and growing recognition of the diversity and importance of single-celled organisms and other "lower" creatures. Most of Western history favored the biblically sanctioned twofold division of organisms into plants and animals, with a third realm for all inorganic substances—leading to the old taxonomy of "animal, vegetable, or mineral" in such venerable games as Twenty Questions.

This twofold division produced a host of practical consequences, including the separation of biological research into two academic departments and traditions of study: zoology and botany. Under this system, all single-celled organisms had to fall into one camp or the other, however uncomfortably, and however tight the shove of the shoehorn. Thus, paramecia and amoebae became animals because they move and ingest food.

Photosynthesizing unicells, of course, became plants. But what about photosynthesizers with mobility? And, above all, what about the prokaryotic bacteria, which bear no key feature suggesting either allocation? But since bacteria have a strong cell wall, and because many species are photosynthetic, bacteria fell into the domain of botany. To this day, we still talk about the bacterial "flora" of our guts.

By the time I entered high school in the mid-1950s, expansion and enlightenment had proceeded far enough to acknowledge that unicells could not be so divided by criteria of the multicellular world and that single-celled organisms probably deserved a separate kingdom of their own, usually called Protista.

Twelve years later, as I left graduate school, even greater respect for the unicells had led to further proliferation at the "lower" end. A "five kingdom" system was now all the rage (and has since become canonical in textbooks), with the three multicellular kingdoms of plants, fungi and animals in a top layer (representing, loosely, production, decomposition and ingestion as basic modes of life); the eukaryotic unicells, or Kingdom Protista, in a middle layer; and the prokaryotic unicells.

Most proponents of this system recognized the gap between prokaryotic and eukaryotic organization—that is, the transition from Monera to Protista—as the fundamental division within life, thus finally granting bacteria their measure of independent respect, if only as a bottom tier.

Starting in the mid-1970s, development of techniques for sequencing the genetic code finally gave us a key for mapping evolutionary relationships among bacterial lineages. We know how to use anatomy for drawing genealogical trees of multicellular creatures more familiar to us. But we are so ignorant of the bacterial world that we couldn't identify proper genealogical divisions, and we therefore tended to dump all bacteria together into a bag of little unicellular blobs, rods and spirals.

As nucleotide sequences began to accumulate for key segments of bacterial genomes, a fascinating and unsuspected pattern emerged and has grown ever stronger with passing years and further accumulation of evidence. This group of supposed primitives, once shoved into one small bag for their limited range of overt anatomical diversity, actually includes two great divisions, each far larger in scope (in terms of genomic distinction and variety) than all three multicellular kingdoms (plants,animals and fungi) combined!

Moreover, one of these divisions seemed to gather together, into one grand sibship, most of the bacteria living in odd environments and working by peculiar metabolisms under extreme conditions (often in the absence of oxygen) that may have flourished early in Earth's history—the methanogens, or methane producers; the tolerators of high salinities, the halophiles; and the thrivers at temperatures around the boiling point of water, the thermophiles.

These first accurate genealogical maps led to the apparently inescapable conclusion that two grand kingdoms, or domains, must be recognized within the old Kingdom Monera—(1) Bacteria, for most conventional forms that come to mind when we contemplate this category (the photosynthesizing blue-greens, the gut bacteria, the organisms that cause human diseases and therefore become "germs" in our vernacular); and (2) Archaea, for the newly recognized coherence of oddballs. By contrast, all eukaryotic organisms, the three multicellular kingdoms as well as all unicellular eukaryotes, belong to a third great evolutionary domain, the Eucarya.

The accompanying chart, adapted from the work of Carl Woese, our greatest pioneer in this new constitution of life, says it all, with the maximally stunning device of a revolutionary picture. We now have a system of three grand evolutionary domains—Bacteria, Archaea and Eucarya—and two of the three consist entirely of prokaryotes: that is, "bacteria" in the vernacular, the inhabitants of life's constant mode. Once we place two-thirds of evolutionary diversity at life's mode, we have much less trouble grasping the centrality of this location and the constant domination of life by bacteria.

    Figure [1]

For example, the domain of Bacteria, as presently defined, contains several major subdivisions, and the genetic distance between any pair is at least equal to the average separation between eukaryotic kingdoms such as plants and animals.

Note, by contrast, the restricted domain of all three multicellular kingdoms. On this genealogical chart for all life, the three multicellular kingdoms form three little twigs on the bush of just one among three grand domains of life. Quite a change in one generation—from my parents' learning that everything living must be animal or vegetable, to the icon of my mature years: the kingdoms Animalia and Plantae as two little twigs amid a plethora of other branches on one of three bushes, with both other bushes growing bacteria, and only bacteria, all over.


The taxonomic criterion, while impressive, does not guarantee bacterial domination—and for a definite reason common to all genealogical schemes. Bacteria form the root of life's entire tree. For the first 2 billion years or so, about half of life's full history, bacteria alone built the tree of life. Therefore, all multicellular creatures, as late arrivers, can only inhabit some topmost branches; the roots and trunk must be exclusively bacterial.

This geometry does not make the case for calling our modern world an "Age of Bacteria" because the roots and trunk might now be atrophied, with only the multicellular branches flourishing. We need to show not only that bacteria build most of life's tree but also that these bacterial foundations remain strong, healthy, vigorous and fully supportive of the minor superstructure called multicellular life. Bacteria, indeed, have retained their predominant position and hold sway not only by virtue of a long and illustrious history but also for abundant reasons of contemporary vigor. Consider two aspects of ubiquity:

1. Numbers. Bacteria inhabit effectively every place suitable for the existence of life. Mother told you, after all, that bacterial "germs" require constant vigilance to combat their ubiquity in every breath and every mouthful, and the vast majority of bacteria are benign or irrelevant to us, not harmful agents of disease. One fact will suffice: during the course of life, the number of E. coli in the gut of each human being far exceeds the total number of people that now live and have ever lived.

Numerical estimates, admittedly imprecise, are a stock in trade of all popular writing on bacteria. The Encyclopaedia Britannica tells us that bacteria live by "billions in a gram of rich garden soil and millions in one drop of saliva." Writer Dorion Sagan and biologist Lynn Margulis write in their book, Garden of Microbial Delights, that "human skin harbors some 100,000 microbes per square centimeter" ("microbes" includes nonbacterial unicells, but the overwhelming majority of "microbes" are bacteria.

I was particularly impressed with their statement about our colonial status: "Fully 10 percent of our own dry body weight consists of bacteria, some of which, although they are not a congenital part of our bodies, we can't live without."

2. Places. Since the temperature tolerance and metabolic ranges of bacteria so far exceed the scope of all other organisms, bacteria live in all habitats accessible to any form of life, while the edges of life's toleration are almost exclusively bacterial—from the coldest puddles on glaciers to the hot springs of Yellowstone Park, to oceanic vents where water issues from the earth's interior at 480 degrees F (still below the boiling point at the high pressures of oceanic bottoms).

At temperatures greater than 160 degrees F, all life is bacterial. Thermophila acidophilum thrives at 140 degrees F, and at a pH of 1 or 2, the acidity of concentrated sulfuric acid. This species, found on the surface of burning coals and in the hot springs of Yellowstone Park, effectively freezes to death below 100 degrees F.


Importance for human life forms the narrowest of criteria for assessing the role of any organism in the history and constitution of life, though the conventional case for bacteria proceeds largely in this mode. I will therefore expand a bit toward utility (or at least "intrinsicness") for all of life and even for the Earth.

1. Historical. Oxygen, the most essential constituent of the atmosphere for human needs, now maintains itself primarily through release by multicellular plants in the process of photosynthesis. The Earth's original atmosphere apparently contained little or no free oxygen, and this otherwise unlikely element both arose historically and is now maintained by the action of organisms.

Plants may provide the major input today, but oxygen started to accumulate in the atmosphere about 2 billion years ago, substantially before the evolution of multicellular plant life. Bacterial photosynthesis supplied the atmosphere's original oxygen and, in concert with multicellular plants, continues to act as a major source of resupply today.

We could not digest and absorb food properly without our gut "flora." Grazing animals, cattle and their relatives, depend upon bacteria in their complex, quadripartite stomachs to digest grasses in the process of rumination. About 30 percent of atmospheric methane can be traced to the action of methanogenic bacteria in the guts of ruminants, largely released into the atmosphere—how else to say it—by belches and farts.

In another symbiosis essential to human agriculture, plants need nitrogen as an essential soil nutrient but cannot use the ubiquitous free nitrogen of our atmosphere. This nitrogen is "fixed," or chemically converted into usable form, by the action of bacteria like Rhizobium, living symbiotically in bulbous growths on the roots of leguminous plants.

2. Current. We could also compile a long list of more parochial uses for human needs and pleasures: the degradation of sewage to nutrients suitable for plant growth; the possible dispersion of oceanic oil spills; the production of cheeses, buttermilk and yogurt by fermentation (we make most alcoholic drinks by fermentation of eukaryotic yeasts); the bacterial production of vinegar from alcohol and of MSG from sugars.

More generally, bacteria (along with fungi) are the main reducers of dead organic matter and thus act as one of the two major links in the fundamental ecological cycle of production (photosynthesis) and reduction to useful form for renewed production. (The ingesting animals are just a little blip upon this basic cycle; the biosphere could do very well without them.) Sagan and Margulis write in conclusion:

"All of the elements crucial to global life—oxygen, nitrogen, phosphorus, sulfur, carbon—return to a usable form through the intervention of microbes. . . . Ecology is based on the restorative decomposition of microbes and molds, acting on plants and animals after they have died to return their valuable chemical nutrients to the total living system of life on Earth."

N E W   D A T A   O N   B A C T E R I A L   B I O M A S S

This range of bacterial habitation and necessary activity certainly makes a good case for domination of life by the modal bacter. But one claim, formerly regarded as wildly improbable but now quite plausible, if still unproven, would really clinch the argument. We may grant bacteria all the above, but surely the main weight of life rests upon eukaryotes, particularly upon the wood of our forests. Another truism in biology has long proclaimed that the highest percentage of the Earth's biomass—pure weight of organically produced matter—must lie in the wood of plants.

Bacteria may be ubiquitous and present in nearly uncountable numbers, but they are awfully light, and you need several gazillion to equal the weight of even a small tree. So how could bacterial biomass even come close to that of the displacing and superseding eukaryotes? But new discoveries in the open oceans and Earth's interior have now made a plausible case for bacterial domination in biomass as well.

Bacteria dwell in virtually every spot that can sustain any form of life. And we have underestimated their global number because we, as members of a kingdom far more restricted in potential habitation, never appreciated the full range of places that might be searched.

For example, the ubiquity and role of bacteria in the open oceans have been documented only in the past 20 years. Conventional methods of analysis missed up to 99 percent of these organisms because we could identify only what could be cultured from a water sample, and most species don't grow on most culture media. Now, with methods of genomic sequencing and other techniques, we can assess taxonomic diversity without growing a large, pure culture of each species.

Scientists had long known that the photosynthesizing Cyanobacteria ("blue-green algae" of older terminology) played a prominent role in the oceanic plankton, but the great abundance of heterotrophic bacteria (nonphotosynthesizers that ingest nutrients from external sources) had not been appreciated. In coastal waters, these heterotrophs constitute from 5 to 20 percent of microbial biomass and can consume an amount of carbon equal to 20 to 60 percent of total "primary production" (that is, organic material made by photosynthesis)—giving them a major role near the base of oceanic food chains.

But Jed A. Fuhrman and his colleagues then studied the biomass of heterotrophic bacteria in open oceans (by far the largest habitat on Earth by area) and found that they dominate in these environments. In the Sargasso Sea, for example, heterotrophic bacteria contribute 70 to 80 percent of microbial carbon and nitrogen and form more than 90 percent of biological surface area.

In the late 1970s, marine biologists discovered the bacterial basis of food chains for deep-sea vent faunas and the unique dependence of this community upon energy from the earth's interior, rather than from a solar source. Two kinds of vents had been described: cracks and small fissures with warm water emerging at temperatures of 40 degrees to 70 degrees F and large conical sulfide mounds, up to 30 feet in height, and spouting superheated waters at temperatures that can exceed 600 degrees F.

Bacteria had long been identified in waters from small fissures of the first category, but it was only in the early 1980s that John Baross and his colleagues discovered a bacterial biota, including both oxidative and anaerobic species, in superheated waters emanating from the sulfide mounds (also known as "smokers").

They cultured bacteria from waters collected at 650 degrees F and then grew vigorous communities in a laboratory chamber with waters heated to 480 degrees F at a pressure of 265 atmospheres. Thus, bacteria can (and do) live in high temperatures (and pressures) of waters flowing beneath Earth's surface.

Then, in the early 1990s, several groups of scientists found and cultured bacteria from oil drillings and other environments beneath oceans and continents, thus indicating that bacteria may live generally in the Earth's interior and not only in limited areas where superheated waters emerge at the surface: from four oil reservoirs nearly two miles below the bed of the North Sea and below the permafrost surface of Alaska's North Slope, from a Swedish bore hole nearly four miles deep and from fourwells about a mile deep in France's East Paris Basin.

Water migrates extensively through cracks and joints in subsurface rocks and even through pore spaces between grains of sediments themselves (an important property of rocks, known as "porosity" and vital to the oil industry as a natural mechanism for concentrating underground liquids—and, as it now appears, bacteria as well). Thus, although such data do not indicate global pervasiveness or interconnectivity of subsurface bacterial biotas, we certainly must entertain the proposition that much of the Earth deep beneath our feet teems with microbial life.

We might ask one further question that would clinch the case for underground ubiquity: Moving away from the specialized environments of deep-sea vents and oil reservoirs, do bacteria also live more generally in ordinary rocks and sediments (provided that some water seeps through joints and pore spaces)? New data from the mid-1990s seem to answer this most general question in the affirmative as well.

R.J. Parkes found abundant bacteria in ordinary sediments of five Pacific Ocean sites at depths up to 1,800 feet. Meanwhile, the Department of Energy, under the leadership of Frank J. Wobber, had been digging deep wells to monitor contamination of groundwater from both inorganic and potentially microbial sources (done largely to learn if bacteria might affect the storage of nuclear wastes in deep repositories!). Wobber's group, taking special pains to avoid the risk of contamination from surface bacteria introduced into the holes, found bacterial populations in at least six sites, including a boring in Virginia at 9,180 feet under the ground!

In 1995, T.O. Stevens and J.P. McKinley described rich bacterial communities living more than 3,000 feet below Earth's surface in rocks of the Columbia River Basalt in the northwestern United States. These bacteria are anaerobic and seem to get energy from hydrogen produced in a reaction between minerals in the basaltic rocks and groundwater seeping through.

Thus, like the biotas of the deep-sea vents, these bacteria live on energy from the Earth's interior, entirely independent of the photosynthetic, and ultimately solar, base of all conventional ecosystems. To confirm their findings in the field, Stevens and McKinley mixed crushed basalt with water free from dissolved oxygen. This mixture did generate hydrogen. They then sealed basalt together with groundwaters containing the deep bacteria. In these laboratory conditions, simulating the natural situation at depth, the bacteria thrived for up to a year.

Following a scientific tradition for constructing humorous and memorable acronyms, Stevens and McKinley have named these deep bacterial floras, independent of solar energy and cut off from contact with surficial communities, SLiME (for subsurface lithoautotrophic microbial ecosystem—the second word is just a fancy way of saying "getting energy from rocks alone"). Jocelyn Kaiser, writing a comment for Science magazine on the work of Stevens and McKinley, used a provocative title: "Can deep bacteria live on nothing but rocks and water?" The answer seems to be yes.

When one considers how deeply entrenched has been the dogma that most earthly biomass lies in the wood of our trees, this potentially greater weight of underground bacteria represents a major revision of conventional biology and quite a boost for the modal bacter.

Not only does the Earth contain more bacterial organisms than all others combined (scarcely surprising, given their minimal size and mass); not only do bacteria live in more places and work in a greater variety of metabolic ways; not only did bacteria alone constitute the first half of life's history, with no slackening in diversity thereafter; but also, and most surprisingly, total bacterial biomass (even at such minimal weight per cell) may exceed all the rest of life combined, even forest trees, once we include the subterranean populations as well.  Need any more be said in making a case for the modal bacter as life's constant center of maximal influence and importance?

[ Stephen Jay Gould, "Planet of the Bacteria," Washington Post Horizon, 1996, 119 (344): H1; Reprinted here with permission; This essay was adapted from Full House, New York: Harmony Books, 1996, pp. 175-192. ]

Home Page  |  Further Reading  |  Site Map  |  Send Feedback