The Great Fossil Enigma

by Simon J. Knell

  Recognizing the power of technology, Müller had his Bonn colleagues parcel up 120 papers produced from the SEM’s use and sent them to the university's administrators. Not long afterward, a new and improved model was installed.

  In Ohio, Walt Sweet built an alliance with a colleague in biology in order to secure that university's first SEM. Installed beneath the university's clock tower, it operated twenty-four hours a day and had its own dedicated staff member. It was by this means independent of any departmental control or politics. Sweet thought this ideal. Chris Barnes in Canada also became active in SEM studies, these helping him understand the relationship between structure and ecology. In 1970, for example, he and his collaborators came to believe that robust Ordovician fossils with a shallow basal cavity and little or no white matter were characteristic of near-shore, hypersaline, carbonate environments.5

  This explosion of activity peaked in the early 1970s. It brought to light a host of new structures and supported studies of evolution, classification, and the mechanical properties of conodont fossils, but it left many of the most fundamental gaps in understanding unresolved. As Rhodes reflected at the time, “The use of scanning electron microscopy has revolutionized our knowledge (though not yet our understanding) of the internal structure and surface architecture of conodonts.”6 As a physical object, by the early 1970s, it certainly seemed that the conodont element was known in every detail, yet the science was still no nearer to understanding how these elements, or indeed the assemblages, functioned, and consequently it had no chance of comprehending the nature of the animal itself. The work of Hass and Gross had been superseded, but these new studies did not immediately displace the constraints they had put on the fossil. Of rather more importance to the future of the animal was a general unshackling of the imagination arising from the cultural shift taking place in the science in the early 1970s that encouraged paleontologists to become freethinkers, modelers, and theorists. It was this that would allow conodont workers to again think creatively about the animal much as Lindström had done in 1964.

  In this respect, Melton and Scott's animal, with its bizarre stomach teeth, had certainly done the science a service because it put forward the most bizarre of explanations for how the conodont elements functioned. Lindström, of course, had been asked to reflect on the nature of the real conodont animal for the meeting at which the Bear Gulch beast was presented to the world. He had to do this without actually knowing much about that animal, but as soon as he saw it, he was perhaps relieved to know that it was not really the conodont animal at all. Indeed, in the book edited by Rhodes in which Melton and Scott's animal was described, he mourned the fact that so little progress had been made. The new statistical work of Bergström and Sweet, and others, seemed far more illuminating, for it had revealed that the animal had only a limited number of assemblage architectures, each linked to the other by an evolutionary thread. The existence of a clear evolutionary relationship between these different assemblage types encouraged Lindström to believe that there must also be a functional relationship. It was logical to consider that they all did roughly the same job, just as the teeth in our mouths do pretty much what they did in the mouths of our ape ancestors and their ancestors before them. This idea – that these elements had particular functions – was also supported by the knowledge that similar architectures and characters evolved repeatedly. It was already being established that shape correlated with lifestyle, and this further enforced the view that function was directly linked to form. This thinking led to the asking of new questions. For example, platforms, which look most like crushing teeth, sit alongside elements lacking a single plane of growth. If these latter elements could not have closed as one tooth does upon another, how could the platforms next to them do so? More to the point, platforms seemed to have evolved from conodonts of this other kind.7 This sort of reasoning suggested even the most tooth-like elements could not have functioned as teeth. Even apparatuses composed entirely of simple conical elements could not, Lindström thought, have functioned as jaws, as the elements were too long, too thin, or simply wrongly shaped.

  With teeth out of the picture, Lindström reflected on those SEM images, which showed a honeycomb pattern of pitting on platform surfaces at right angles to the main denticle row. These looked similar to muscle attachment pits, encouraging him to imagine the conodont elements supporting tentacles. Fused clusters were now of more use to him; perhaps these conodont elements had been brought together through the contraction of muscles on death. Using this kind of reasoning, Lindström refined his view of the oral cavity of the conodont animal, picturing “a wormlike, soft-bodied animal, moving or floating in most cases freely in the sea.” At five to fifty millimeters long, he reasoned that “it probably fed on microplankton that it strained from the sea-water by means of a lophophore, which was supported by phosphatic elements and may have formed a ring round the mouth” (figure 12.2b).

  Lindström's biological thinking had, since his arrival in Marburg, been greatly affected by weekly lunches with distinguished algal biologist Hans-Adolf von Stosch. He also began to spend two or three weeks a year with his students at a marine station in Roscoff, Brittany. There, surrounded by an outstanding library, he looked in detail at the full range and morphology of marine life through the eyes of a conodontologist. With these increasingly biologically informed thoughts in his mind, the final stage in the development of his personal interpretation of the animal resulted from an invitation from the International Palaeontological Association to give its prestigious annual address at Burlington House in London in March 1974. After several decades of serious but muted geological research, the newsworthy sensation of Melton and Scott's beast had returned the conodont animal to its rightful place as paleontology's greatest and best-loved enigma. Behind this sensation, however, stood a period of rapid progress that had stabilized the language, revealed an advanced understanding of the animal's evolution and microstructure, and proven the fossil's unrivaled stratigraphic utility. As Lindström spoke it would have been clear to this diverse and informed paleontological audience that he possessed arcane knowledge, that conodont studies possessed their own complexity, sophistication, and expertise. Lindström would also have come across as a man who delighted in the thought experiment and lateral thinking. As a speaker he was doubtless aware of how compelling such skills become when attempting to unravel a seemingly insoluble puzzle.

  Lindström challenged his audience to imagine the functional interrelationships between the different fossilized parts of the animal, each of which was in a process of evolutionary remodeling. Surely they could see, as Lindström could, that the morphology of the elements, with denticles arranged along different planes, meant they could not be teeth. At the time, the arrow worm or chaetognath was a popular model for the conodont. Seddon and Sweet had used it in their ecological model as a behavioral indicator, and Siegfried Rietschel, at the Senckenberg Museum in Frankfurt, having compared conodont assemblages with the radulae of gastropods, jaws of polychaete worms, and grasping apparatus of chaetognaths, thought the latter provided the best anatomical model for those conodont animals possessing simple cone-like elements. Rietschel admitted that this model could not be extended to more complex forms, but Lindström thought the argument fundamentally flawed: Conodonts fossils were often too thin, curved, or complex to have been used for seizing or holding food. He was supported in his doubts by the new evidence of a fossil chaetognath found in the Pennsylvanian Mazon Creek Lagerstätte in Illinois.8 Here, iron-rich nodules, exposed during strip mining and left on heaps of spoil, were found to contain a multitude of fossil worms, including an arrow worm looking very much like its modern-day cousins. Although its head was poorly preserved, the find suggested that chaetognaths had coexisted with, rather than evolved from, the conodont animal.

  Lindström now adjusted and developed his original model, attempting to accommodate all that was securely known and believed. There was no better person to do this. Lindström wa
s considered a consummate conodont connoisseur – even Ziegler tipped his hat to him. Nevertheless, there had to be some art to clothing bones in flesh; this could not be achieved through objective science alone. Lindström reasoned that the structure of the conodont element must have had a bearing on the fleshy structures attached to it. Conodont evolution showed that the surface area of elements increased over time, but what function, he asked himself, would warrant this? Breathing, excretion, and the uptake of nutrients – these all seemed possibilities, but only the latter had specific requirements of shape. If tentacles were attached to muscles on the surface of the elements, they could be extended and retracted. The sharp and pointed faces of the elements were, he reasoned, pointed away from each other because, covered in flesh, they could not have articulated one against the other without causing damage. This, he thought, was reflected in the assemblages that had been found.

  He considered other possibilities but settled on a barrel-shaped organ. It was still a lophophore. For Lindström, Melton and Scott's conodont-eating animal – as he saw it – limited the overall size of the conodont animal to seven to nine millimeters and constrained its shape: “The animal is more likely to have been oblong or even barrel-shaped rather than long and worm shaped.” Now the animal emerged: “The conodont animal thus sketched was not necessarily very mobile. It might even have been a passive floater, relying on its battery of unpalatable denticles for its protection. Some conodont animals may have formed colonies…. The food gathered by conodonts might have been both microscopic particulate matter and dissolved material.” This animal reflected Lindström's awareness of modern forms of life possessing lophophores and an earlier view that conodonts and brachiopods might have a common ancestor.9 With the animal and its lifestyle out in the open, he placed it in a real-world environment: “If this is true we might expect the conodonts to occur most plentifully in areas where such nutrients are abundant, as for instance in environments characterized by upwelling deeper ocean water. Such environments may occur on the margins of oceanic troughs or along submarine rises. The occurrence of conodonts in certain fossil sediments (black muds and trough-rise limestone facies) appears to agree with this prediction.” Lindström's “barrel” or “gooseberry” was now established, a sophisticated reading of the evidence, clothed in flesh and dropped in the ocean (figure 12.2c).

  12.2. Lindström's imagined animals: (a) the conodont lophophore is born in 1964; (b) it developed to fringe the oral cavity of the animal in 1973; and (c) in 1974, this became Lindström's barrel. Reproduced with permission from M. Lindström, Conodonts (1964), Elsevier; M. Lindström in Conodont Paleozoology (1973), and M. Lindström, Palaeontology 17 (1974).

  The paper demonstrated Lindström's mastery of his field, but he left certain things unsaid. It had occurred to him that the animal forming in his mind looked something like a spiny ostracod. Perhaps the ostracod was mimicking the conodont animal with its armory of tough spines. With this thought his mind, the conodont animal acquired its barrel – or ostracod-like – shape. It was an idea to which he intended to give further thought, but when he did so, it occurred to him there were no predators in that sea capable of seeing with any precision an animal so small. So what use could there be for this kind of mimicry? He laughed to himself and forgot about it. He remained attached to the idea that the conodont elements might support tentacles on a lophophore, but for him the barrel was a rather more ephemeral imagining. It continued to exist, however, in the published paper, there to be the subject of serious debate and occasional ridicule (as Lindström's “toilet roll”) long after that thought had, for Lindström, passed.

  Simon Conway Morris was among those who took Lindström's barrel seriously as he pondered a curious fossil from another – the most famous – Lagerstätte: the Middle Cambrian Burgess Shale of British Columbia, Canada.10 He had chanced upon a sawn slab and put it to one side while searching through the huge Burgess Shale collections at the National Museum of Natural History in Washington. A short while later its counterpart turned up. Conway Morris, who was to build his career on bringing to light the weird and wonderful Burgess Shale fauna, knew this fossil was something special: “The specimen had evidently never been noted by any other worker. No other specimens have been found.” Poorly preserved as a thin film darker than the surrounding rock, this gelatinous, flattened, worm-like animal was a contemporary of the earliest known conodonts. It lacked a distinct head but possessed a curious, almost figure-eight-shaped apparatus within which could be detected the impressions of twenty-five “teeth” very similar in shape to conodont fossils of that period. Unfortunately, most of what had caused these impressions had leached away – a common problem with calcareous fossils in the shale. Nevertheless, Conway Morris thought them conodont-like in size and shape and noted that they too possessed a basal depression and showed some variation in form around the apparatus. He thought this reflected a possible “symmetry transition,” an idea from Lindström's book, which Conway Morris used as his main guide to the conodonts. The counterpart specimen lacked these teeth but instead seemed to show the decayed remnants of tentacles of a kind seen in some other rare fossils. But unlike paleontologists studying common fossils, Conway Morris's science lacked the firm footing of familiarity. He had to decide what constituted informative data on the two specimens and what might have resulted from post-death compaction and alteration. His interpretations depended on his own connoisseurship of fossils and on the work of others – such as Lindström's barrel and transition series, and a 1932 interpretation of tentacles – work he could not know well and relied upon the interpretations of others. These were his building blocks for imagining the animal; in effect, Lindström offered him both spectacles and blinkers. He could not have imagined the animal he now drew without them.

  12.3. Conway Morris's animal. The conodont elements fringing the mouth supported a tentacle-covered lophophore. This lophophore had been reasoned into existence by Maurits Lindström. Reproduced with permission from S. Conway Morris, Palaeontology 19 (1976).

  Conway Morris produced an animal that swam by contracting muscles along its sides much like a modern nemertine worm (figure 12.3). The “teeth” were not really teeth. They were too fragile to produce a rasping lophophore, but they may have supported tentacles on a lophophore covered in fine cilia, which directed food toward the mouth. He looked for a modern-day animal group with this same feeding structure and found it in brachiopods, the colonial bryozoans that resemble tiny corals, and a group of sessile worms that live in chitinous tubes. These animals confirmed to Conway Morris that his fossil did possess a lophophore, but they also told him that his animal, which he called Odontogriphus omalus (from “toothed riddle”), was not related to any known type of lophophore-possessing creature.

  Agreeing with Lindström, Conway Morris dismissed Melton and Scott's animal as a predator of the conodont animal. Now Conway Morris believed he possessed the conodont animal. It was rather larger than Lindström had imagined, but Conway Morris felt Melton and Scott's predator could accommodate it. And while Lindström's animal floated, Conway Morris argued that modern animals of similar size possess locomotion. He also did not like the way Lindström had placed his lophophore on the exterior of the animal; he thought this improbable, but he accommodated this by imagining Lindstrom's more complex lophophore as an evolutionary development on that seen in his own animal. By these means, Conway Morris gently shoehorned the nose of his worm onto Lindström's barrel, claiming without fanfare to have found a conodont animal – an animal that could not be allied to any existing group. He saw his animal as ancestral to the conodont animals that blossomed in the Paleozoic and, so he thought from reading Lindström's book, continued up to the Cretaceous. And now a recent discovery in Kazakhstan seemed to take the conodont even further back in time, to the boundary with the Pre-Cambrian.11

  Gould later remarked on Conway Morris's discovery: “What a potential coup for a beginner – to discover the secret of secrets, and resolve a ce
ntury of debate!” Whether Gould was ever really convinced by this animal is uncertain; most conodont workers were not. This was, however, a science of improbabilities, so ably demonstrated a few years later by a little debate that emerged on the pages of Science concerning “conodont pearls.” Tiny phosphatic spheres up to 0.7 millimeters in diameter, each possessing a dimple, had been found in their thousands in the United States and Australia, associated with conodonts and composed of the same layered structure. Brian Glenister, Gil Klapper, and Karl Chauff thought they resulted from irritation caused by detritus or parasites.12 They were, it seemed, the equivalent of the oyster's pearl and perhaps akin to the balls associated with Müller's Westergaardodina. Duncan McConnell and David Ward at Ohio State University were unconvinced, reasoning that these spheres were the nautilus-equivalent of bladder stones, well known in the modern animal. Glenister and company stood their ground but the topic faded.