3. Part–counterpart. When you split a rock to find a fossil, you get two for the price of one—the fossil itself (called the part) and the impression of the organism forced into layers above (called the counterpart)—thumb and thumbprint, if you will. The part, as the actual fossil, has been favored by scientists and collectors; the counterpart, as an impression, has less to offer in traditional evaluations. Walcott worked almost exclusively with parts, and frequently didn’t bother to keep the counterparts at all. (When he did collect counterparts, he often didn’t catalogue them with the matching parts. They ended up in different drawers or relegated to the spoil heaps of less interesting material. Some he even gave away in trade with other museums.)
For a traditional fossil, coherently made of a single piece—the shell of a clam or snail, for example—the distinction between part and counterpart is obvious. The specimen is the part; the mold on the upper surface, the counterpart. Under Walcott’s view of Burgess organisms as single films, the same clear difference applies—the film itself is the part; its impression, the less interesting counterpart.
But when Whittington revealed the three-dimensional nature of the Burgess fossils, this easy distinction and differential rating disappeared. An arthropod contains hundreds of articulating pieces; since these are preserved on several adjacent layers in the Burgess Shale, splitting a rock at a bedding plane cannot yield a clear division, with the entire organism (the part) on one surface, and only the impression (the counterpart) on the other. Any split must leave some pieces of the organism on one side, other bits on the opposite block. In fact, the distinction between part and counterpart ultimately breaks down for the Burgess fossils. You can only say that one surface preserves more interesting anatomy than the other. (By convention, the Burgess workers finally decided to designate the top view upon the organism as the part, and the view looking up as the counterpart. By this scheme, for an animal like Sidneyia, eyes, antennae, and other features of the external carapace are often preserved on the counterpart, legs and internal anatomy on the part.)
All expeditions from 1966 to the present have rigorously collected both part and counterpart (when preserved), keeping and cataloguing them together. Some of the greatest Burgess discoveries of the past twenty years have occurred at the Smithsonian when a Walcott counterpart, sometimes uncatalogued, sometimes even classified in a different phylum, was recognized and reunited with its part. Can you top this for a heart-warming tale, more satisfying (since less probable) than the reunion of Gabriel with Evangeline? In 1930, the Raymond expedition found a specimen of Branchiocaris pretiosa, an exceedingly rare arthropod with fewer than ten known examples. In 1975 (when Derek Briggs had already submitted his monograph on this species for publication), the Royal Ontario Museum expedition found the counterpart of this specimen, still lying on the talus slope in British Columbia where Raymond and his party had spurned it forty-five years before!
Obviously, if both part and counterpart contain important bits of anatomy, we must study them together if we strive for tolerable completeness in reconstruction. (In their camera lucida drawings, Whittington and colleagues have followed the convention of including information from both part and counterpart in the same figure.) Reassociation of part with counterpart has resolved a puzzle in the study of Sidneyia. Based on an isolated specimen, Walcott had suggested a peculiar reconstruction for the gills of Sidneyia. But Bruton, examining both Walcott’s part and the “counterpart which Dr. D. E. G. Briggs observantly found among uncatalogued material in the Walcott Collection” (Bruton, 1981, p. 640), discovered that the supposed gill did not belong to Sidneyia at all. Conway Morris later identified this fossil as a decayed and folded specimen of the priapulid worm Ottoia prolifica.
3.9. Camera lucida drawing of a specimen of Sidneyia preserved in an unusual orientation. We are looking at the front end head on, and therefore can appreciate the convexity of the animal—information that we cannot get in the usual orientation. Note in particular the positions of insertion for the antennae (labeled Ra and La) and for the eye (e).
3.10. A specimen of Sidneyia in an unusual orientation that reveals the arrangement of the legs. We are looking head on at a cross section through the front end of the body, just behind the head, and can see the first four legs on the animal’s right side, compressed together (labeled Rl1–Rl4). The alimentary canal (al), in the center of the body, is also visible.
3.11. Camera lucida drawing of a walking leg of Sidneyia. Note the strong spines (labeled gn, for “gnathobase”) at the point of insertion for the leg into the body. This array of spines bordering the food groove suggests that the animal was a predator. The leg is so well preserved that we can count the segments and infer the orientation in life.
These three procedures—excavation, odd orientations, and part-counterpart—are guides to the three dimensional reanimation of squashed and distorted fossils. They don’t tell us much about other aspects of life among Burgess organisms—how they moved and ate, and how they grew, for example. Unfortunately, for all its virtues in preserving anatomy, the Burgess Shale, as a transported assemblage buried in a mud cloud, does not provide other kinds of evidence that more conventional faunas often include. We have no tracks or trails, no burrows, no organisms caught in the act of eating their fellows—in short, few signs of organic activity in process. For some reason not well understood (and most unfortunately), the Burgess Shale includes almost no juvenile stages of organisms.
Still, some procedures beyond those already noted have been useful in particular cases; they will be discussed in turn as the organisms enter our story. I have already mentioned the gut contents of Sidneyia. Other organisms have also been identified as carnivores by a study of their alimentary tracts. For example, in the gut of a priapulid worm Conway Morris found smaller members of the same species—the world’s earliest example of cannibalism—and numerous hyolithids. He also used varying degrees of decay to resolve the anatomy of the priapulid worm Ottoia prolifica. Bruton (for Sidneyia, Leanchoilia, and Emeraldella) and Briggs (for Odaraia) made three-dimensional models from composites of their drawings and photographs. Conway Morris has used injuries and patterns of growth to understand the habits of the enigmatic Wiwaxia. He argues (1985) that in a unique Burgess example of growth caught in the act, one specimen was buried in the process of molting—casting off an old garment for an entirely new outer coat of plates and spines.
THE CHRONOLOGY OF A TRANSFORMATION
What do scientists “do” with something like the Burgess Shale, once they have been fortunate enough to make such an outstanding discovery? They must first perform some basic chores to establish context—geological setting (age, environment, geography), mode of preservation, inventory of content. Beyond these preliminaries, since diversity is nature’s principal theme, anatomical description and taxonomic placement become the primary tasks of paleontology. Evolution produces a branching array organized as a tree of life, and our classifications reflect this genealogical order. Taxonomy is therefore the expression of evolutionary arrangement. The traditional medium for such an effort is a monograph—a descriptive paper, with photographs, drawings, and a formal taxonomic designation. Monographs are almost always too long for publication in traditional journals; museums, universities, and scientific societies have therefore established special series for these works. (As noted before, most Burgess descriptions have appeared in monographs published by the Royal Society of London in their Philosophical Transactions—a series for long papers.) These monographs are expensive to produce and have strictly limited circulation, mostly to libraries.
This situation has engendered the unfortunate condescension expressed toward monographs and their authors by many scientists from other disciplines. These works are dismissed as exercises in “mere description,” a kind of cataloguing that could as well be done by clerks and drones. At most, some credit may be given for care and attention to detail—but monographs do not emerge as the vanguard of creative novelty.
Some monographs are pedestrian, of course—the description of a new brachiopod or two from a well-known formation deposited during the heyday of the group’s success will raise few eyebrows—but then a great deal of workaday physics and chemistry is also dial-twirling to iterate the obvious. The best monographs are works of genius that can transform our views about subjects inspiring our passionate interest. How do we know about Lucy, the “ape-man of Java,” our Neanderthal cousins, the old man of Cro-Magnon, or any of the other human fossils that fire our imagination as fully as an Apollo landing on the moon, except by taxonomic monographs? (Of course, in these cases of acknowledged “newsworthiness,” highly touted preliminary reports long precede any technical publication, usually providing, as the cliché goes, much heat with little light.)
* * *
TAXONOMY AND THE STATUS OF PHYLA
The world is so full of a number of things,
I’m sure we should all be as happy as kings.
—Robert Louis Stevenson
This famous couplet, from A Child’s Garden of Verses, expresses the chief delight of our natural world and the primary result of evolution—incredible and irreducible variety. Since the human mind (in its adult version, at least) craves order, we make sense of this variety by systems of classification. Taxonomy (the science of classification) is often undervalued as a glorified form of filing—with each species in its folder, like a stamp in its prescribed place in an album; but taxonomy is a fundamental and dynamic science, dedicated to exploring the causes of relationships and similarities among organisms. Classifications are theories about the basis of natural order, not dull catalogues compiled only to avoid chaos.
Since evolution is the source of order and relationship among organisms, we want our classifications to embody the cause that makes them necessary. Hierarchical classifications work well in support of this aim because the primary topology of life’s tree—the joining of twigs to branches, branches to limbs, and limbs to trunks as we trace species back to ever earlier common ancestors—can be expressed by a system of ever more inclusive categories. (People join with apes and monkeys to make primates; primates with dogs to make mammals; mammals with reptiles to make vertebrates; vertebrates with insects to make animals, and so on. Since Linnaeus and other pre-Darwinians also used hierarchical systems, evolution is not the only possible source of order expressed by this form; but evolution by diversification does imply branching from common ancestry, and such a topology is best rendered by hierarchical classification.)
Modern taxonomies recognize seven basic levels of increasing inclusion—from species (considered as the fundamental and irreducible units of evolution) to kingdoms (the broadest groupings of all): species, genera, families, orders, classes, phyla, and kingdoms.
At the highest level—the kingdom—the old folk division into plants and animals, and the old schoolboy system of plants, animals, and single-celled protists, have been largely superseded by a more convenient and accurate five-kingdom system: Plantae, Animalia, and Fungi for multicellular organisms; Protista (or Protoctista) for single-celled organisms with complex cells; and Monera for single-celled organisms (bacteria and cyanophytes) with simple cells devoid of nuclei, mitochondria, and other organelles.
The next level—the phylum—is the basic unit of differentiation within kingdoms. Phyla represent the fundamental ground plans of anatomy. Among animals, for example, the broadest of basic groups are designated as phyla—sponges, “corals” (including hydras and jellyfish), annelids (earthworms, leeches, and marine polychaetes), arthropods (insects, spiders, lobsters, and the like), mollusks (clams, snails, squid), echinoderms (starfishes, sea urchins, and sand dollars), and chordates (vertebrates and their kin). In other words, phyla represent the major trunks of life’s tree.
This book treats the early history of the animal kingdom. In focusing on the origin of phyla and their early number and degree of differentiation, we ask the most basic of all questions about the organization of our animal kingdom.
How many phyla of animals does our modern earth contain? Answers vary, since this question involves some subjective elements (a terminal twig is an objective thing, and species are real units in nature, but when is a branch large enough to be called a bough?). Still, we note some measure of agreement; phyla tend to be big and distinct. Most textbooks recognize between twenty and thirty animal phyla. Our best modern compendium, a book explicitly dedicated to the designation and description of phyla (Margulis and Schwartz, 1982) lists thirty-two animal phyla—a generous estimate in comparison with most. In addition to the seven familiar groups already mentioned, the animal phyla include, among others, the Ctenophora (comb jellies), Platyhelminthes (flatworms, including the familiar laboratory Planaria), Brachiopoda (bivalved invertebrates common as Paleozoic fossils, but rarer today), and Nematoda (unsegmented roundworms, usually tiny and fantastically abundant in soil and as parasites).
After such a long disquisition, the point of this exegesis with respect to the Burgess Shale may be quickly stated: the Burgess Shale, one small quarry in British Columbia, contains the remains of some fifteen to twenty organisms so different one from the other, and so unlike anything now living, that each ought to rank as a separate phylum. We hesitate to give such a “high” designation to single species because our traditions dictate that phyla achieve their distinctness through hundreds of speciation events, each building a bit of the total difference, piece by piece. Hence, the anatomy of a group should not become sufficiently distinct to rank as a separate phylum until a great deal of diversity has been accumulated by repeated speciation. According to this conventional view—obviously incorrect or incomplete by evidence from the Burgess—lineages of one or a few species cannot diverge far enough to rank as phyla. But que faire? The fifteen to twenty unique Burgess designs are phyla by virtue of anatomical uniqueness. This remarkable fact must be acknowledged with all its implications, whatever decision we ultimately make about the formalities of naming.
* * *
The worst of human narrowness pours forth in the negative assessment of monographic work as merely descriptive. Scientific genius is equated with an oddly limited subset of intellectual activities, primarily analytical ability and quantitative skill, as though anyone could describe a fossil but only the greatest thinkers could conceive of the inverse-square law. I wonder if we will ever get past the worst legacy of IQ theory in its unilinear and hereditarian interpretation—the idea that intelligence can be captured by a single number, and that people can be arrayed in a simple sequence from idiot to Einstein.
Genius has as many components as the mind itself. The reconstruction of a Burgess organism is about as far from “simple” or “mere” description as Caruso from Joe Blow in the shower, or Wade Boggs from Marvelous Marv Throneberry. You can’t just look at a dark blob on a slab of Burgess shale and then by mindless copying render it as a complex, working arthropod, as one might transcribe a list of figures from a cash-register tape into an account book. I can’t imagine an activity further from simple description than the reanimation of a Burgess organism. You start with a squashed and horribly distorted mess and finish with a composite figure of a plausible living organism.
This activity requires visual, or spatial, genius of an uncommon and particular sort. I can understand how this work proceeds, but I could never do it myself—and I am therefore relegated to writing about the Burgess Shale. The ability to reconstruct three-dimensional form from flattened squashes, to integrate a score of specimens in differing orientations into a single entity, to marry disparate pieces on parts and counterparts into a functional whole—these are rare and precious skills. Why do we downgrade such integrative and qualitative ability, while we exalt analytical and quantitative achievement? Is one better, harder, more important than the other?
Scientists learn their limitations and know when they need to collaborate. We do not all have the ability to assemble wholes from pieces. I once spent a week in the field with Richard Leakey, and
I could sense both his frustration and his pride that his wife Meave and their coworker Alan Walker could take tiny fragments of bone and, like a three-dimensional jigsaw puzzle, put together a skull, while he could do the work only imperfectly (and I saw nothing at all but fragments in a box). Both Meave and Alan showed these skills from an early age, largely through a passion for jigsaw puzzles (curiously, both, as children, liked to do puzzles upside down, working by shapes alone, with no help from the picture).
Harry Whittington, who shares this rare visual genius, also expressed his gift at an early age. Harry began with no particular advantages of class or culture. He grew up in Birmingham, the son of a gunsmith (who died when Harry was only two) and grandson of a tailor (who then raised him). His interests wandered toward geology, thanks largely to the inspiration of a sixth-form (just pre-university) geography teacher. Yet Harry had always recognized and exploited his skill in three-dimensional visualization. As a child, he loved to build models, mostly of cars and airplanes, and his favorite toy was his Meccano set (the British version of an Erector set, providing strips of steel that can be bolted together into a variety of structures). In beginning geology courses, he excelled in map interpretation and, especially, in drawing block diagrams. The consistent theme is unmistakable: a knack for making three-dimensional structures from two-dimensional components, and inversely, for depicting solid objects in plane view. This ability to move from two to three dimensions, and back again, provided the key for reconstructing the fauna of the Burgess Shale.