Wonderful Life: The Burgess Shale and the Nature of History Read online

  Darwin, honest as always in exposing the difficulties of his theory, placed the Cambrian explosion at the pinnacle of his distress, and devoted an entire section to this subject in the Origin of Species. Darwin acknowledged the anti-evolutionary interpretation of many important geologists: “Several of the most eminent geologists, with Sir. R. Murchison at their head, are convinced that we see in the organic remains of the lowest Silurian* stratum the dawn of life on this planet” (1859, p. 307). Darwin recognized that his theory required a rich Precambrian record of precursors for the first complex animals:

  If my theory be true, it is indisputable that before the lowest Silurian stratum was deposited, long periods elapsed, as long as, or probably far longer than, the whole interval from the Silurian age to the present day; and that during these vast, yet quite unknown periods of time, the world swarmed with living creatures (1859, p. 307).

  Darwin invoked his standard argument to resolve this uncomfortable problem: the fossil record is so imperfect that we do not have evidence for most events of life’s history. But even Darwin acknowledged that his favorite ploy was wearing a bit thin in this case. His argument could easily account for a missing stage in a single lineage, but could the agencies of imperfection really obliterate absolutely all evidence for positively every creature during most of life’s history? Darwin admitted: “The case at present must remain inexplicable; and may be truly urged as a valid argument against the views here entertained” (1859, p. 308).

  Darwin has been vindicated by a rich Precambrian record, all discovered in the past thirty years. Yet the peculiar character of this evidence has not matched Darwin’s prediction of a continuous rise in complexity toward Cambrian life, and the problem of the Cambrian explosion has remained as stubborn as ever—if not more so, since our confusion now rests on knowledge, rather than ignorance, about the nature of Precambrian life.

  Our Precambrian record now stretches back to the earliest rocks that could contain life. The earth is 4.5 billion years old, but heat from impacting bodies (as the planets first coalesced), and from radioactive decay of short-lived isotopes, caused our planet to melt and differentiate early in its history. The oldest sedimentary rocks—the 3.75-billion-year-old Isua series of west Greenland—record the cooling and stabilization of the earth’s crust. These strata are too metamorphosed (altered by heat and pressure) to preserve the morphological remains of living creatures, but Schidlowski (1988) has recently argued that this oldest potential source of evidence retains a chemical signature of organic activity. Of the two common isotopes of carbon, 12C and 13C, photosynthesis differentially uses the lighter 12C and therefore raises the ratio of isotopes—12C/13C—above the values that would be measured if all the sedimentary carbon had an inorganic source. The Isua rocks show the enhanced values of 12C that arise as a product of organic activity.*

  Just as chemical evidence for life may appear in the first rocks capable of providing it, morphological remains are also as old as they could possibly be. Both stromatolites (mats of sediment trapped and bound by bacteria and blue-green algae) and actual cells have been found in the earth’s oldest unmetamorphosed sediments, dating to 3.5–3.6 billion years in Africa and Australia (Knoll and Barghoorn, 1977; Walter, 1983).

  Such a simple beginning would have pleased Darwin, but the later history of Precambrian life stands strongly against his assumption of a long and gradual rise in complexity toward the products of the Cambrian explosion. For 2.4 billion years after the Isua sediments, or nearly two-thirds of the entire history of life on earth, all organisms were single-celled creatures of the simplest, or prokaryotic, design. (Prokaryotic cells have no organelles—no nucleus, no paired chromosomes, no mitochondria, no chloroplasts. The much larger eukaryotic cells of other unicellular organisms, and of all multicellular creatures, are vastly more complex and may have evolved from colonies of prokaryotes; mitochondria and chloroplasts, at least, look remarkably like entire prokaryotic organisms and retain some DNA of their own, perhaps as a vestige of this former independence. Bacteria and blue-green algae, or cyanophytes, are prokaryotes. All other common unicellular organisms—including the Amoeba and Paramecium of high-school biology labs—are eukaryotes.)

  The advent of eukaryotic cells in the fossil record some 1.4 billion years ago marks a major increment in life’s complexity, but multicellular animals did not follow triumphantly in their wake. The time between the appearance of the first eukaryotic cell and the first multicellular animal is longer than the entire period of multicellular success since the Cambrian explosion.

  The Precambrian record does contain one fauna of multicellular animals preceding the Cambrian explosion, the Ediacara fauna, named for a locality in Australia but now known from rocks throughout the world. But this fauna can offer no comfort to Darwin’s expectation for two reasons. First, the Ediacara is barely Precambrian in age. These animals are found exclusively in rocks just predating the explosion, probably no more than 700 million years old and perhaps younger. Second, the Ediacara animals may represent a failed, independent experiment in multicellular life, not a set of simpler ancestors for later creatures with hard parts. (I shall discuss the nature and status of the Ediacara fauna in chapter V.)

  In one sense, the Ediacara fauna poses more problems than it solves for Darwin’s resolution of the Cambrian explosion. The most promising version of the “imperfection theory” holds that the Cambrian explosion only marks the appearance of hard parts in the fossil record. Multicellular life may have undergone a long history of gradually ascending complexity leaving no record in the rocks because we have found no “Burgess Shale,” or soft-bodied fauna, for the Precambrian. I would not challenge the contribution of this eminently sensible argument to the resolution of the Cambrian enigma, but it cannot provide a full explanation if Ediacara animals are not ancestors for the Cambrian explosion. For the Ediacara creatures are soft-bodied, and they are not confined to some odd enclave stuck away in a peculiar Australian environment; they represent a world-wide fauna. So if the true ancestors of Cambrian creatures lacked hard parts, why have we not found them in the abundant deposits that contain the soft-bodied Ediacara fauna?

  Puzzles mount upon puzzles the more we consider details of the astounding 100-million-year period between the Ediacara fauna and the consolidation of modern body plans in the Burgess Shale. The beginning of the Cambrian is not marked by the appearance of trilobites and the full range of modern anatomy identified as the Cambrian explosion. The first fauna of hard parts, called the Tommnotian after a locality in Russia (but also world-wide in extent), contains some creatures with identifiably modern design, but most of its members are tiny blades, caps, and cups of uncertain affinity—the “small shelly fauna,” we paleontologists call it, with honorable frankness and definite embarrassment. Perhaps efficient calcification had not yet evolved, and the Tommnotian creatures are ancestors that had not yet developed full skeletons, but only laid down bits of mineralized matter in small and separate places all over their bodies. But perhaps the Tommotian fauna is yet another failed experiment, later supplanted by trilobites and their cohort in the final pulse of the Cambrian explosion.

  Thus, instead of Darwin’s gradual rise to mounting complexity, the 100 million years from Ediacara to Burgess may have witnessed three radically different faunas—the large pancake-flat soft-bodied Ediacara creatures, the tiny cups and caps of the Tommotian, and finally the modern fauna, culminating in the maximal anatomical range of the Burgess. Nearly 2.5 billion years of prokaryotic cells and nothing else—two-thirds of life’s history in stasis at the lowest level of recorded complexity. Another 700 million years of the larger and much more intricate eukaryotic cells, but no aggregation to multicellular animal life. Then, in the 100-million-year wink of a geological eye, three outstandingly different faunas—from Ediacara, to Tommotian, to Burgess. Since then, more than 500 million years of wonderful stories, triumphs and tragedies, but not a single new phylum, or basic anatomical design, added to the Burgess
complement.

  Step way way back, blur the details, and you may want to read this sequence as a tale of predictable progress: prokaryotes first, then eukaryotes, then multicellular life. But scrutinize the particulars and the comforting story collapses. Why did life remain at stage 1 for two-thirds of its history if complexity offers such benefits? Why did the origin of multicellular life proceed as a short pulse through three radically different faunas, rather than as a slow and continuous rise of complexity? The history of life is endlessly fascinating, endlessly curious, but scarcely the stuff of our usual thoughts and hopes.

  LIFE AFTER THE BURGESS: SOFT-BODIED FAUNAS AS WINDOWS INTO THE PAST

  An old paleontological in joke proclaims that mammalian evolution is a tale told by teeth mating to produce slightly altered descendant teeth. Since enamel is far more durable than ordinary bone, teeth may prevail when all else has succumbed to the whips and scorns of geological time. The majority of fossil mammals are known only by their teeth.

  Darwin wrote that our imperfect fossil record is like a book preserving just a few pages, of these pages few lines, of the lines few words, and of those words few letters. Darwin used this metaphor to describe the chances of preservation for ordinary hard parts, even for maximally durable teeth. What hope can then be offered to flesh and blood amidst the slings and arrows of such outrageous fortune? Soft parts can only be preserved, by a stroke of good luck, in an unusual geological context—insects in amber, sloth dung in desiccated caves. Otherwise, they quickly succumb to the thousand natural shocks that flesh is heir to—death, disaggregation, and decay, to name but three.

  And yet, without evidence of soft anatomy, we cannot hope to understand either the construction or the true diversity of ancient animals, for two obvious reasons: First, most animals have no hard parts. In 1978, Schopf analyzed the potential for fossilization of an average modern marine fauna of the intertidal zone. He concluded that only 40 percent of genera could appear in the fossil record. Moreover, potential representation is strongly biased by habitat. About two-thirds of the sessile (immobile) creatures living on the sea floor might be preserved, as contrasted with only a quarter of the burrowing detritus feeders and mobile carnivores. Second, while the hard parts of some creatures—vertebrates and arthropods, for example—are rich in information and permit a good reconstruction of the basic function and anatomy of the entire animal, the simple roofs and coverings of other creatures tell us nearly nothing about their underlying organization. A worm tube or a snail shell implies very little about the organism inside, and in the absence of soft parts, biologists often confuse one for the other. We have not resolved the status of the earth’s first multicellular fauna with hard parts, the Tommotian problem (discussed in chapter V), because these tiny caps and covers provide so little information about the creatures underneath.

  Paleontologists have therefore sought and treasured soft-bodied faunas since the dawn of the profession. No pearl has greater price in the fossil record. Acknowledging the pioneering work of our German colleagues, we designate these faunas of extraordinary completeness and richness as Lagerstätten (literally “lode places,” or “mother lodes” in freer translation). Lagerstätten are rare, but their contribution to our knowledge of life’s history is disproportionate to their frequency by orders of magnitude. When my colleague and former student Jack Sepkoski set out to catalogue the history of all lineages, he found that 20 percent of major groups are known exclusively by their presence in the three greatest Paleozoic Lagerstätten—the Burgess Shale, the Devonian Hunsrückschiefer of Germany, and the Carboniferous Mazon Creek near Chicago. (I shall, for the rest of this book, use the standard names of the geological time scale without further explanation. If you spurn, dear reader, my exhortation to memorize this alphabet, please refer to figure 2.1. I also recommend the mnemonics at the beginning of this chapter.)

  An enormous literature has been generated on the formation and interpretation of Lagerstätten (see Whittington and Conway Morris, 1985). Not all issues have been resolved, and the ins and outs of detail provide endless fascination, but three factors (found in conjunction only infrequently) stand out as preconditions for the preservation of soft-bodied faunas: rapid burial of fossils in undisturbed sediment; deposition in an environment free from the usual agents of immediate destruction—primarily oxygen and other promoters of decay, and the full range of organisms, from bacteria to large scavengers, that quickly reduce most carcasses to oblivion in nearly all earthly environments; and minimal disruption by the later ravages of heat, pressure, fracturing, and erosion.

  As one example of the Catch-22 that makes the production of Lagerstätten so rare, consider the role of oxygen (see Allison, 1988, for a dissenting view on the importance of anoxic habitats). Environments without oxygen are excellent for the preservation of soft parts: no oxidation, no decay by aerobic bacteria. Such conditions are common on earth, particularly in stagnant basins. But the very conditions that promote preservation also decree that few organisms, if any, make their natural home in such places. The best environments therefore contain nothing to preserve! The “trick” in producing Lagerstätten—including the Burgess Shale, as we shall see—lies in a set of peculiar circumstances that can occasionally bring a fauna into such an inhospitable place. Lagerstätten are therefore rooted in rarity.

  If the Burgess Shale did not exist, we would not be able to invent it, but we would surely pine for its discovery. The Good Lord of Earthly Reality seldom answers our prayers, but he has come through for the Burgess. If Aladdin’s djinn had appeared to any paleontologist before the discovery of the Burgess, and stingily offered but one wish, our lucky beneficiary would surely have said without hesitation: “Give me a soft-bodied fauna right after the Cambrian explosion; I want to see what that great episode really produced.” The Burgess Shale, our djinn’s gift, tells a wonderful story, but not enough for a book by itself. This fauna becomes a key to understanding the history of life by comparison with the strikingly different pattern of disparity in other Lagerstätten.

  Rarity has but one happy aspect—given enough time, it gets converted to fair frequency. The discovery and study of Lagerstätten has accelerated greatly in the past ten years, inspired in part by insights from the Burgess. The total number of Lagerstätten is now large enough to provide a good feel for the basic patterns of anatomical disparity through time. If Lagerstätten were not reasonably well distributed we would know next to nothing about Precambrian life, for everything from the first prokaryotic cells to the Ediacara fauna is a story of soft-bodied creatures.

  As its primary fascination, the Burgess Shale teaches us about an amazing difference between past and present life: with far fewer species, the Burgess Shale—one quarry in British Columbia, no longer than a city block—contains a disparity in anatomical design far exceeding the modern range throughout the world!

  Perhaps the Burgess represents a rule about the past, not a special feature of life just after the Cambrian explosion? Perhaps all faunas of such exquisite preservation show a similar breadth of anatomical design? We can only resolve this question by studying temporal patterns of disparity as revealed in other Lagerstätten.

  The basic answer is unambiguous: the broad anatomical disparity of the Burgess is an exclusive feature of the first explosion of multicellular life. No later Lagerstätten approach the Burgess in breadth of designs for life. Rather, proceeding forward from the Burgess, we can trace a rapid stabilization of the decimated survivors. The magnificently preserved, three-dimensional arthropods from the Upper Cambrian of Sweden (Müller, 1983; Müller and Walossek, 1984) may all be members of the crustacean line. (As a result of oddities in preservation, only tiny arthropods, less than two millimeters in length, have been recovered from this fauna, so we can’t really compare the disparity in these deposits with the story of larger-bodied Burgess forms.) The Lower Silurian Brandon Bridge fauna from Wisconsin, described by Mikulic, Briggs, and Kluessendorf (1985a and 1985b), contains (like the Burg
ess) all four major groups of arthropods. It also includes a few oddballs—some unclassifiable arthropods (including one creature with bizarre winglike extensions at its sides) and four wormlike animals, but none so peculiar as the great Burgess enigmas like Opabinia, Anomalocaris, or Wiwaxia.

  The celebrated Devonian Hunsrückschiefer, so beautifully preserved that fine details emerge in X-ray photos of solid rock (Stürmer and Bergström, 1976 and 1978), contains one or two unclassifiable arthropods, including Mimetaster, a probable relative of Marrella, the most common animal in the Burgess. But life had already stabilized. The prolific Mazon Creek fauna, housed in concretions that legions of collectors have split by the millions over the past several decades, does include a bizarre wormlike animal known as the Tully Monster (officially honored in formal Latin doggerel as Tullimonstrum). But the Burgess motor of invention had been shut off by then, and nearly all the beautiful fossils of Mazon Creek fit comfortably into modern phyla.

  When we pass through the Permo-Triassic extinction and come to the most famous of all Lagerstätten—the Jurassic Solnhofen limestone of Germany—we gain enough evidence to state with confidence that the Burgess game is truly over. No fauna on earth has been better studied. Quarrymen and amateur collectors have been splitting these limestone blocks for more than a century. (These uniform, fine-grained stones are the mainstay of lithography, and have been used, almost exclusively, for all fine prints in this medium ever since the technique was invented at the end of the eighteenth century.) Many of the world’s most famous fossils come from these quarries, including all six specimens of Archaeopteryx, the first bird, preserved with feathers intact to the last barbule. But the Solnhofen contains nothing, not a single animal, falling outside well-known and well-documented taxonomic groups.