By: Gould, S.J., Natural History, Jul92, Vol. 101, Issue 7


The discovery of an enormous underground organism raises some basic questions about individuality


When an animal achieves disembodied immortality by becoming a verb, human speakers usually honor its behavior: we hawk our wares, gull or buffalo our naive competitors, hound our adversaries, and clam up in the face of adversity; we have also been known to man the barricades and kid around with our companions. But plants and other rooted creatures do not feature so great a range of overt actions, and our botanically based verbs therefore tout growth and appearance as sources of metaphor.


Consider the two most prominent examples, citing comparable phenomena but with such different meanings-for one usually expresses our joy and the other our fear. Art and prosperity “flower”; taxes and urban violence “mushroom.” The burden of difference reflects an obvious source in our culture and legends. We love the bright flowers of “higher” plants, either radiant in the sunlight or jewellike in the quiet darkness of a forest. We loathe the spongy, fruiting bodies of “lowly” fungi, growing in dank dampness, sprouting in cancerous formlessness from rotting logs. (Even a colorful mushroom usually strikes us as sinister rather than lovely.) I well remember a common schoolyard taunt, often cruelly directed at unloved classmates: “There's a fungus among us”--a cry that always inspired the ritualistic retort: “Kill it before it multiplies.”


Combine this image of uncontrolled and noxious fungal growth with our general fascination for superlatives-the biggest and most persistent in this case and we can easily understand the flurry of press commentary that greeted a technical article by M. L. Smith, J. N. Bruhn, and J. P. Anderson in the April 2, 1992, issue of Nature: “The fungus Armillaria bulbosa is among the largest and oldest living organisms.” On the same day, the New York Times reported the discovery in a front page story with a more sprightly headline: “Thirty-Acre Fungus Called World's Largest Organism.” [Editor's note: As this column was being prepared for press, the Associated Press reported the find of a larger fungus, an Armillaria ostoyae, covering about two and a half square miles in the state of Washington. Larger organisms probably will be found, but the discoveries will not change the point of this piece.]


We divide muticellular life into three great kingdoms: animals, plants, and fungi. How does this new Armillaria stack up against the champions of the other domains. Blue whales, by far the largest animals that ever lived (edging out both Ultrasaurus and Supersaurus from the world of dinosaurs), may reach 100 feet and as many tons. (Incidentally, since female whales exceed males in size, the largest individual animal of all time was undoubtedly female.) Sequoia trees are bigger by far, sometimes exceeding 1,000 tons (although mostly in nonliving wood), not to mention heights measured in hundreds of feet and years in thousands.


Armillaria bulbosa lives in and around tree roots in European and eastern North American mixed hardwood forests. Clones begin as a single fertilized spore and then spread out by vegetative growth. The basic unit of spread is a hypha, or thread-like filament that forms the structural unit of growth in many fungi. In Armillaria, the hyphae are then bundled into cordlike accumulations called rhizomorphs. Since these rhizomorphs grow and extend underground (in and around tree roots), a human observer sees nothing of this interwoven subterranean mat except for the occasional and spatially discontinuous mushrooms that poke through the forest floor.


Since we mistake an individual mushroom body for a discrete organism, we might look at the area of this Armillaria clone and, seeing nothing of the underground continuity, view the species as consisting of a few widely scattered, entirely separate items-a population of several insignificant individuals. But Smith and his colleagues, working in a forest near Crystal Falls on the Upper Peninsula of Michigan, found a region of some thirty acres underlain by the interconnected rhizomorphs of a single Armillaria clone. Extrapolating from best assessments of rhizomorph growth rates, they infer a minimal age of 1,500 years-probably a substantial underestimate since neighboring clones have inhibited further spread for some time, while the calculation assumes continuous expansion. They estimate the weight of rhizomorphs in the soil at about ten tons, but this figure excludes much of the clone's bulk growing in places harder to assess-for example, within tree roots and deep in the soil. Moreover, this figure excludes the mass of fine hyphae that extend from the rhizomorphs into surrounding soil and wood. Smith and his colleagues estimate that the weight of the entire clone may exceed their figure by ten times or more, yielding a total closer to 100 tons.


Thus, this champion Armillaria clone may weigh as much as a blue whale (although substantially less than a large sequoia) and probably lies in the ballpark of the oldest tree for age. Its claim to “firstest with the mostest” must rest upon the added criterion of lateral spread-in other words, of mushrooming. Thirty acres, however thinly distributed in spots, makes quite a creature. Even the Blob, of B-movie fame, will have to take a back seat.


As so often happens in the first flurry of commentary on scientific findings, the wrong aspects get emphasized and the truly interesting conceptual issues fall into the background. I can get as het up as the next guy about claims for maximal age, size, and weight. I know that Robert Wadlow nearly reached nine feet, that Robert Earl Hughes exceeded 1,000 pounds, and that Mr. Izumi of Japan lived past his 120th birthday. The Michigan clone of Armillaria, while failing short in weight, and perhaps in age, does enter my mental list as the biggest organic spread. But the deeper fascination of this tale lies elsewhere-in the striking way that this underground fungal mat forces us to wrestle with the vital biological (and philosophical) question of proper definitions for individuality.


First of all, the novelty of Smith, Bruhn, and Anderson's study lies not in the discovery of something so big but in the establishment of criteria to test the clone's status as an individual. Large and continuous underground mats of fungal rhizomorphs have long been recognized. Heretofore, however, we had no way of ascertaining whether or not they formed from a single source. Fungal spores are everywhere (as we all know from our experiences with bread mold). A single mushroom can produce up to a million fertile spores per hour over several days. With so many spores, representing so many different clones, falling all over the forest floor, why suspect that a large mat of rhizomorphs should represent growth from one initiating spore, and therefore count as a single individual by genealogy? Why not propose that such gigantic mats of rhizomorphs are actually congeries, or aggregations, made of products grown from several founding spores (representing many different parents), all twisted and matted together-in other words, a heap rather than a person?


Spores are certainly ubiquitous, but a constant rain does not guarantee a cooperative matting together to form massive aggregations, for organisms have immunological mechanisms for distinguishing self from nonself and rejecting amalgamation (abutting coral colonies do not generally fuse on growing reefs), while Darwinian theory views competition among different genetic units within a species as a major thrust of life's game. Thus, single versus multiple sources for large fungal mats has been viewed as an open and interesting question for some time.


Genetic tests are now available for resolving such an issue. Smith and colleagues first sampled several portions of the Armillaria mat both for genes that determine compatibility in breeding (called mating-type alleles) and for several fragments of mitochondrial DNA. Both the mating-type and the mitochondrial genes are highly variable within the species Armillaria bulbosa, but all samples of the Michigan mat yielded the same array. This demonstrates close genetic relationship but does not prove individuality, for how can we know whether the mat is made from many highly inbred siblings or from a single founder'? An additional genetic test then indicated a single founder. Repeated close inbreeding leads to a marked reduction of genetic variability among offspring-a presumed “deep reason” behind our various laws and taboos against incestuous marriage. Genes that are variable within an individual (called heterozygous because the material and paternal copies differ) will tend to become uniform in offspring if that individual mates with very close relatives. Smith and colleagues traced markers of several DNA sequences from heterozygous genes throughout the mat and found no reduction in their variability-a sign that the entire mat has grown from a single source and does not represent an incestuous amalgamation of numerous sibling subclones.


But this elegant demonstration that the Michigan mat formed from a single source only opens the more interesting and portentous issue of defining individuality the central question, as we shall see, for applying Darwinian theory to nature. The Michigan Armillaria mat grew vegetatively from a single source, but does it qualify as an individual under our usual vernacular definitions? Clive Brasier addressed this question in a commentary that accompanied Smith, Bruhn, and Anderson's original article:


The suggestion of Smith et al. that [the Michigan mat] deserves recognition as one of the largest living organisms, rivaling the blue whale or the giant redwood, invites closer scrutiny. The blue whale and redwood exhibit relatively determinate growth within a defined boundary, whereas fungal mycelia do not.


In other words, a whale is a whale, with flippers and tail, but the Michigan mat just spreads. Moreover, we cannot be sure that the Michigan mat is truly continuous throughout. Pieces may break off and become physically separated from the main mass; but a whale's flipper, if broken off, is dead meat, not a miniwhale. Brasier concludes, with justice:


So although [the Michigan mat's] reputation as a champion genotype [discrete genetic system I may yet be secure, its status as a champion organism depends upon one's interpretation of the rules.


What, then, shall we accept as definitive of individuality, and why is this issue important to biological theory, not merely a verbal game? A single genetic origin followed by growth in physical continuity might represent a good first stab, but the Michigan mat raises a cardinal issue also illustrated by many other creatures, including grass blades and bamboo stalks (although the problem may only represent an unfortunate parochialism based upon our unfair extrapolation of what we know about the character of our own bodies to a false criterion for all life).


Our canonical individuals are bounded entities with definable form-a whale, a tree, a cockroach, a human being. By analogy, we wish to label a grass blade, a bamboo stalk, or a mushroom as an individual. But consider the proposed criterion of contiguity and unique genetic origin. A bamboo stalk looks like what we call an entire plant in other circumstances, but each stalk in a field may arise from a common system of underground runners, all united and all formed by vegetative growth from a single seed. Isn't the bamboo stalk then just an organ of this larger individual, as the mushrooms of Armillaria are only the visible fruiting bodies of an underground totality? (Bamboo stalks are giant grass blades, so the same argument applies to our lawns.)


Botanists, more often than zoologists, must deal with this problem of apparent organs that look like individuals (although a colony of coral animals raises exactly the same issue). Botanists have therefore devised a special terminology to treat these ambiguous cases of parts that are entire organisms in vernacular usage but organs of a larger totality by genetic definition. They speak of the morphologically well-defined part (the grass blade, bamboo stalk, or mushroom) as a ramet and the entire interconnected system (the underground runners and stalks, the mat of rhizomorphs with occasional mushroom buds) as a genet. In other words, the vernacular individual is a ramet; the genetic individual, a genet. This terminology does not solve the conceptual problem of defining individuality but merely devises names to acknowledge the classical case of inherent ambiguity.


The vernacular and genetic definitions can be driven even further apart by recognizing that even the chief character of connectedness can fail when we advocate the genetic criterion. In 1977, zoologist Dan Janzen wrote a provocative article on this subject in the American Naturalist, with the subheading: “What Is an Aphid?” Many aphid species breed sexually just once a year. Females born from this sexual union produce subsequent all-female generations by parthenogenesis-that is, without fertilization. The parthenogenetic offspring are all identical, both to their sisters and their mothers (except for rare new mutations). In other words, they form a clone of genetically identical bodies-and billions of aphids may ultimately arise from one stem-mother. (Eventually, the females begin to produce some male offspring and fertilization can then ensue but this is a different story.)


Now what are all these aphid bodies in a single parthenogenetic clone? By any vernacular criterion, they are individuals. They look like any other unambiguously individual insect-like a cockroach, a ladybug, or a cricket. They have mouths and six legs, just like any ordinary insect. Yet they are part of one genetic system, all formed by an analogue of vegetative growth from a single starting point (the stem-mother). The entire clone of separated aphid bodies is just like the stalks of a bamboo field or the mushrooms of an Armillaria mat, but without the underground connections. The aphid bodies are disconnected ramets of a single genet. Why, Janzen argues, should we not label the aphid bodies as pans and the entire clone as a single EI, or “evolutionary individual”? This redefinition has some startling consequences, as Janzen notes (you may need to read this quotation twice to put yourself in its unconventional framework, but the implications are surely arresting):


Each EI grows rapidly by parthenogenesis, with occasional pieces (aphids) being bitten out of it by parasites (in conventional discussions these would be called ... predators). Only very rarely is an EI preyed upon (i.e., all of it eaten), since part of its growth pattern is to spread itself very thinly over the surface of the plants in its habitat, so thinly that a potential predator is very unlikely to find all of it at once.


Before this sequence of increasing ambiguity becomes any more maddening, let me propose a different approach, and a potential solution. Terms are best defined within the context of explanatory theories. Gravity may have a variety of vernacular meanings, but its changing technical definition in the successive formulations of Newton and Einstein define its character as a scientific concept. Similarly, individual has a central role and meaning within Darwinian theory, our ruling paradigm for understanding nature. Shouldn't we accept this theoretical definition as our primary biological meaning? (The vernacular may stray, but science cannot and should not control all ordinary usage.)


Darwin's central postulate states that natural selection works upon individuals engaged in a struggle (metaphorical and without conscious intent to be sure) for reproductive success. Individuals that leave more surviving offspring obtain a Darwinian edge, and populations change thereby. Fine, but who is the “individual” engaged in such a struggle? Darwin gives us a clear answer: individuals are organisms-that is, conventional bodies (with some nuancing for ambiguous cases like mushrooms and aphids). Natural selection works on creatures-on individual lions fighting for a limited supply of zebras; on individual trees struggling with others for light.


This emphasis on ordinary vernacular organisms was central to Darwin's radical reformulation of nature, for he was consciously striving to overthrow the comforting and conventional idea of nature's intrinsic benevolence, with a creator fashioning good organic design and harmonious ecosystems directly. How delicious to contemplate that these “benevolent” results arise only as side consequences of a mechanism operating “below” divine superintendence, and having no “goal” but the selfish propagation of individuals that is, organisms struggling for personal reproductive success and nothing else.


We continue to accept Darwin's abstract formulation today-individuals struggling for personal reproductive success-but a substantial rethinking has enlarged Darwin's concept of individuality. For Darwin, only organisms are individuals, or “units of selection.” But what properties does an entity need to operate as a Darwinian individual-and are organisms the only such entities? We can specify five such properties: an individual must have a clear beginning (or birth) point, a clear ending (or death) point, and sufficient stability between to be recognized as an entity. These first three properties suffice to define an individual in the most abstract sense. But an entity requires two more properties to enter a Darwinian process of reproductive competition: it must bear children, and these children must be produced by a principle of inheritance that makes offspring resemble parents with the possibility of some differences.


Darwin was surely right that ordinary organisms (also Armillaria mats and aphid clones) possess these five properties: they are born and die at definable points; they are stable enough during their lifetimes; and they have children that resemble them with the possibility of difference. Organisms can therefore be units of selection.


But what about entities either more or less inclusive than organisms? What about genes “below” organisms, and species “above”? We usually think of genes as parts and species as collectivities, but maybe this conventional view only represents the bias of restricted focus on our personal lives. Maybe genes and species are just as good Darwinian individuals as bodies. After all, species are born when a population becomes isolated and branches off from its parental stock. Species die unambiguously at their extinction (unless of course the species slowly evolves into one or more other species!jdDec1202). Most species are quite stable throughout their geological duration. Genes also possess the five key properties of birth, death, stability, reproduction, and inheritance with the possibility of difference.


Thus, individuality extends beyond Armillaria mats and aphid clones to encompass different levels of biological organization-so different that we have usually called them parts or collectivities under the parochial assumption that only organisms can be individuals. Genes and species are also Darwinian individuals, and selection can operate upon them as well. Selection does not only sort organisms, as Darwin thought; it operates simultaneously at several levels of a genealogical hierarchy-on genes and cell lineages “below” organisms, and on populations and species “above” organisms. All these levels of the hierarchy produce legitimate Darwinian individuals-and this is the large, inclusive, and proper biological sense of the term individual.


Evolutionary theory does not operate as Darwin imagined when selection acts simultaneously upon several kinds and levels of individuals. Balances and feedback, rather than adaptive perfection, become the source of temporary stabilities. Fancy feathers are wonderful for an individual peacock but harmful (in geological time) for the individual species Pavo cristatus. Roger Clemens's salary is terrific for him and his family line but of dubious value for long-term persistence of the individual called Major League Baseball.


Nature is not an intrinsic harmony of clearly defined units. Nature is built by multiple levels, interacting fuzzily at their borders. We cannot even formulate an unambiguous definition of individual at the single level of organic bodies-as Armillaria mats and aphid clones demonstrate. Moreover, in Darwinian terms, legitimate individuals exist and operate at several levels of a genealogical hierarchy-genes and species, as well as organisms. But what a fascination when this maelstrom of differing individuals forms its meshwork of interaction to produce life's history by Darwinian evolution. Does nature herself then sing Walt Whitman's Song of Myself?


Do I contradict myself?

     Very well then I contradict myself,

(I am large, I contain multitudes.)