The figure that faces the principal entrance is the most remarkable in this excavation, and has given rise to numberless conjectures and theories. It is a gigantic bust, representing some three-headed being, or three heads of some being to whom the temple may be supposed to be dedicated.
CAPTAIN BASIL HALL,
Fragments of Voyages and Travels (1832)
THREE
The Animal with Three Heads
IN 1933, TED BRANSON AND MAURICE MEHL BELIEVED THE conodont would remain forever silent on the question of its anatomy. But they were wrong. Indeed, at the very moment they took possession of the fossil and turned it into a geological abstraction, new discoveries were being made that threatened to tear their utilitarian dream apart. These discoveries did not do so, however, because Branson and Mehl's bubbling pots of mud and practical science fit perfectly into a country infatuated with oil. Who, by comparison, really cared about the biology of a tiny, obscure creature? Who would willingly sacrifice the fossils’ usefulness for the sake of incorporating this new anatomical information? Carey Croneis, doyen of the new micropaleontology at the University of Chicago, certainly valued this practical turn, but he objected to the willingness of oil company geologists to sacrifice science for the sake of economic gain. He felt that the very integrity of the new science was at stake and called upon the industry to employ “men not only of adequate scholastic attainments but even more important, men of a high type of intellectual potentiality, which is, of course, a very different thing.” His was not a solitary voice, but the economic reality of the new industrial paleontology was never going to be affected by the moralizing of paleontologists in universities and museums. Ted Branson's son, Carl, for example, working for Shell in Texas in the late 1940s, revealed how fundamentally different this utilitarian world was: “It has been five years since I have seen many non-oil seekers; too long…. I'm mostly tied to hunting for grease and get no time for reading or research.”1 As a result, in the United States, two overlapping cultures developed around microfossils. One was committed wholly to the economic project. For it, fossils were no more than abstract tools, and biological concepts, such as evolution, simply devices to be used to distinguish as many unique “species” (or time markers) as possible. The other community also valued the practical benefits of fossils, but it saw the fossils embedded more properly in sciences that sought to understand the past conditions of the earth and life upon it. One group, fed on its greasy diet, soon grew obese in participants, while the other remained small and, since it trained the new oil men and women, could never fully separate itself from the practical science. For many types of fossil this division of labor caused few problems because the fossils themselves were simple objects. The conodont, however, was a biological mystery and it was, as we shall see, about to acquire considerable complexity. This produced an animal with a schizophrenic identity.
The development of this divided world was further aided by Branson and Mehl's belief that they were the pioneers of a new science. This encouraged them to think that no one else was doing equivalent science. With their rapid conquest of the fossil and its distribution, their work assumed an intellectual independence and authority blind to what was going on further afield. But they were not alone in this. The geological community in American universities and state surveys in the 1920s and 1930s was small and incestuous. Through networks of friends, students, teachers, and other contacts it sought to distribute opportunities so that no one was really in competition and everyone had their own patch. In a country so vast, this encouraged particular outlooks to develop in geographically isolated research communities. At the University of Chicago and the Illinois State Geological Survey, for example, the science never became as utilitarian as it did in Missouri. At each of these two centers, Croneis and Branson, respectively, shaped minds and projects and defined what, for their students, should be considered the new science of microfossils.
We have already explored one side of this polarized world in Branson and Mehl's successful attempt to realize the fossils’ practical worth. The other side of this science sought to reveal more about the animal itself and it did so over the same period. As we shall see, this division of labor resulted in two incompatible truths and three distinct and irreconcilable animals. To understand how this came about, we must return to the beginning of the twentieth century and think not about rocks and oil but about the animal.
In America, in the early twentieth century, so Clinton Stauffer tells us, Karl von Zittel and Josef Victor Rohon's worm held sway: “This conclusion, although perhaps favoring the minority, was rather generally accepted by workers throughout America, and conodonts were classified with the Vermes.” However, this worm was destined to have an ephemeral existence. Was it not inevitable that a new generation would look at these fossils and see the fish others had seen? E. O. Ulrich and Ray Bassler believed the conodont a fish in 1915. A decade later, their comparison of conodont and fish fossils in the U.S. National Museum convinced them that each type of conodont belonged to a separate species of fish, as we have seen. This was, of course, the invention of a convenient truth that chimed with Ulrich's belief that geologists should concern themselves only with the usefulness of fossils, not with their biology.2
Others were also remaking the conodont fish, and doing so several years before Ulrich and Bassler published their classification. Among them was William Bryant. It was he who had taken Hinde's Polygnathus name and restricted its meaning to the conodont's distinctive “crested crushing plates,” soon to be called “platforms.” These tiny objects seemed to speak to him – though they did so silently – of a past that was alive: “In upper Devonian times a grinding dentition was introduced, bearing mute witness to the struggle for existence of these creatures as the conditions of life became more difficult.” In Hinde's remarkable Conodont Bed at Eighteen-mile Creek, which Bryant considered “one of the richest and most interesting storehouses of ancient vertebrate life to be found anywhere in the world,” he found evidence to show that the animal had later acquired dermal armor. By 1921, a beast was emerging, in his imagination at least, and it was not a worm: “On the whole, the longer I have studied these organisms, the more have I become convinced that the true Conodonts have hardly anything really diagnostic in common with Annelid jaws. If, as I shall hereinafter try to demonstrate, certain of the leaf-like forms are of the nature of pavement teeth, then the conclusion seems almost unavoidable that the Conodonts must be considered as the dentition of some primitive type of fishes.” As he stared at these teeth, he noticed they were paired, left and right; they had come from a bilaterally symmetrical animal.3
Being a specialist in fossil fishes, it was perhaps inevitable that Bryant would see the fossils’ piscine qualities. But he was not alone. At that same moment, John Muirhead Macfarlane, the recently retired professor of botany and modernizing director of the Botanic Garden at the University of Pennsylvania, was wandering some distance from his main subject. Following Hinde's lead, he also believed conodonts to be types of jawless fish related to the hagfish and lamprey (the cyclostomes) and, perhaps influenced by Zittel and Rohon's findings, thought the fossils to be part carbonate of lime and part “horny.” To demonstrate this relationship, he copied Hinde's trick of constructing a striking visual argument, this time illustrating the “teeth” of cyclostomes and conodonts next to each other, without giving any sense of scale.4
Macfarlane then took Hinde's cyclostome and combined it with an old and unpopular idea first published by Ambrosius Hubrecht in 1883. Hubrecht had proposed that vertebrates originated from proboscis or ribbon worms (known today as nemerteans), marine animals capable of injecting a spiked proboscis into their victims’ bodies. Hubrecht suggested that the proboscis, if retained within the body of the animal, could have evolved into the vertebrate backbone. By uniting these ideas – all of which had been sleeping for thirty to forty years – Macfarlane could contend that conodonts “may all represent evolved and complex derivatives from t
he horny teeth formed in the mid part of the proboscis of metanemerteans. For if we accept that freshwater metanemerteans may have swarmed in Cambrian lakes and swamps, these by progressive change may have given rise to Cyclostomes, some of which retained horny teeth as in existing types, while others may have advanced to a more complex calcareous type.”5
This simple piece of interpretive addition seemed to him convincing, and he extended it still further. He noted that the flesh of cyclostomes is rich in oil. If conodonts were similarly oily and had existed in the huge numbers claimed by John Newberry and others, then, he wondered, could they be the source of petroleum? A number of recent biochemical studies encouraged him to think the idea reasonable. So he asked himself in what circumstances might these fish become oil? For this he needed to understand the geology of those rocks that produced oil. He found three well-known American examples that suggested to him an elaborate theory involving earthquakes and volcanoes. These latter, he said, produced rock dust, which fell into and jellified local water bodies, killing off fish populations wholesale. The corpses of these fish were then enveloped in this dust, with the chemical interaction between dust and dead body leading to the formation of oil. The theory had a poetic logic, but little more than that. One reviewer was distinctly unimpressed: “This is an ingenious philosophy, combining all the weaknesses of the organic and inorganic theories of oil origin, with few of the merits of either, and, moreover, an excellent example of the danger of generalizing on two or three particular occurrences. This is a book to look into, but difficult to take seriously.”6
Throughout the 1920s, American paleontologists became increasingly convinced that conodonts were fish. This was aided by the ease with which fish teeth and conodont fossils could be confused. Rohon had wrestled with this problem more than thirty years earlier and it had not gone away: If fish workers were predisposed to see conodonts as fish, then might not conodont workers interpret fish remains as conodonts? In 1928, Chalmer Cooper found what he thought were giant conodonts and sent them to the National Museum for an expert opinion. Paul Roundy, who Cooper thought always too slow to produce results, was impressed but doubtful. He told Cooper, “As regards the ‘jaw’ which you sent, I am not yet prepared to make a definitive statement. Two of them are evidently long, single units with many sharp teeth, and resemble a gigantic specimen of Lonchodus? lineatus. They, however, have several points of difference and my present opinion is that they are not conodont but are true fish remains.”7 Girty then passed Cooper's specimens to fish paleontologist Louis Hussakof for study. Cooper, frustrated at not seeing his material for nearly three years, asked Croneis, who was then his doctorate supervisor, to seek its return. A year later, Ted Branson had somehow received these Oklahoma specimens. Over the preceding period they had remained a well-kept secret, for Branson had not heard about them: “Some material from Oklahoma received today has what seems to be conodonts 40mm long. Our largest heretofore are about 2mm.”8 Given the often frosty relations between Branson and Cooper, it is hard to believe that this material had been sent by the Oklahoma man. Cooper did, however, confer widely, eventually settling on the belief that this material came from fishes. His conclusion denied the existence of a giant conodont but nevertheless seemed to confirm that conodonts were indeed fish.9
What appeared to be final proof that this was the case appeared in 1929, when Stuart Raeburn Kirk of the University of Manitoba published his examination of the Harding Formation at Cañon City in Colorado. Here, nearly forty years earlier, Charles Walcott had famously found “the first definite Ordovician fish remains.” Kirk could now show that the conodonts didn't just occur in the same beds as these fish but that “they show basal attachment to fragments of plates, identical in composition with the fish plates which are so abundantly scattered through the various beds of the Harding.”10 These fish plates belonged to ancestors of those strange, armor-plated jawless fish, known as the ostracoderms, that had flourished in the Devonian.
Kirk was convinced by this association, just as he had been convinced by Ulrich and Bassler's insubstantial evidence for the conodont fish. However, he knew that conodonts had been found in rocks formed long after these fish had become extinct. Stauffer and Helen Plummer thought this a fatal flaw in Kirk's argument and, having found similar conodont material themselves, remained unconvinced that the plates really were composed of bone. Branson and Mehl, however, who also found this bony material, thought it was “the only evidence of the fish nature of conodonts thus far discovered.”11 They certainly did not take Ulrich and Bassler's word for it.
The conodont fish that had been so often denied now began to swim freely in American minds. For example, when Frank Gunnell discovered that a Mesozoic hagfish fossil was intermediate in form between the modern hagfish and the ancient conodont, he could not help but apply a naïve evolutionary determinism then prevalent in U.S. paleontology to connect the dots. He thought it likely that conodonts would one day be found in quite recent rocks, and knowing that the hagfish was considered a primitive vertebrate, he was sure that conodonts would “aid in establishing many heretofore unknown evolutionary relations among the vertebrates.”12 To later scientists such views might seem prophetic, but like his restatement of the fish-petroleum theory, they simply reflected a contemporary moment when these ideas floated unauthored and available for application in that broad church of the new and practically minded micropaleontology.
In America at this time, what might be considered true or known could so easily get mixed up with mere conjecture. Thus when Stauffer and Plummer immersed bar conodonts in hydrochloric acid and discovered that the tooth-like points (denticles) fell out, they could not help but imagine them as tiny jaws with vertebrate-like teeth set in them. Stauffer had visualized such jaws even before he could induce his “teeth” to fall out. The fish that swam through American minds placed expectations into the way these scientists looked. They imagined the fish. Indeed, the enigmatic nature of the animal was once again permitting science to dream and believe, as Macfarlane did, that one day the animal would be found: “It may be hoped however, either that more definite light will be shed by further study of conodonts, or that some layer of subaquatic volcanic ash may yet be discovered, in which as with the medusae, the annelids, and the skates of the Solenhofen slates, fossilized cyclostomes may be discovered.”13
These confident assertions about the conodont fish traveled across the Atlantic to Germany. There, in 1928, near Osterode in the Harz Mountains, work began on a dam, the Sösetalsperre, which was to produce the country's largest water reservoir. It was along the forest road built for the construction traffic that Wilhelm Eichenberg, of the Geological Institute at the nearby University of Göttingen, found great numbers of conodonts on the surfaces of Lower Carboniferous shales.14 His first inclination was to use this material to emulate the detailed microscopic analysis of Pander, Zittel, and Rohon, but his conodonts were extraordinarily brittle and simply shattered when he attempted to expose them. He was familiar with the work of Ulrich and Bassler, Kirk, Hinde, and others, and accepted unquestioningly the American position; conodonts were the remains of fish.
Fortunately for Eichenberg, his university possessed an important expert in these animals; zoologist Franz Stadtmüller was then busy investigating the gill apparatus of living fishes. Stadtmüller gave Eichenberg access to the zoological collections but also advised him on his finds. Eichenberg could now see that conodonts were “skin teeth, plate teeth, plates,” and – a new interpretation – “filter extensions of the gillarches.” In other words, each type of conodont performed a particular function in a single animal and in so doing formed complex respiratory and feeding apparatuses. Following the suggestion of another Göttingen colleague, Hermann Schmidt, Eichenberg introduced the collective term Conodontophorida, or “carriers of the Conodonts,” to refer to this group of animals. By implication, then, “conodonts” were now the “elements” (a zoological term he borrowed from Stadtmüller) making up these appara
tuses. For Pander, “conodonts” were fish, but now Eichenberg made the term apply to the individual fossils. In doing so, he adopted what was already becoming common practice in the United States. His final piece of rethinking involved the naming of the animal itself. Since the elements came from the same stratum and were components in a single animal, as he imagined it, he grouped them together and called the animal that possessed them Prioniodus hercynicus. This name came from just one of the elements, and Eichenberg's use of it in this way was simply an application of the rules of zoological nomenclature, which said that when once separately named fossils are found to belong to a single animal, the longest-established name in that group of fossils should become the name for the animal and thus for all the fossils of which it is composed.
Although based on the importation of American ideas, Eichenberg's analysis meant that Ulrich and Bassler's simple taxonomy, which had yet to take hold, was merely an abstraction, and he called for it to be abandoned. Optimistically, Eichenberg felt the answer to the long-contemplated question of the conodont animal's identity was almost in sight: “All that remains still to be wanting is into which order of fossil fish are the Conodontophorida…to be arranged.”
The fossils making up Hinde's Polygnathus had been found in close association, but that animal proved unacceptable. What chance of survival had Eichenberg's complicated Prioniodus? His animal was even less certain and, being published in German, had other obstacles to overcome before it could penetrate American circles. Now two radically different fish swam in the minds of paleontologists – one separated from the other by the Atlantic. They could coexist because they were separated by numerous physical, linguistic, ideological, and cultural barriers.
The Great Fossil Enigma Page 9