The Great Fossil Enigma

by Simon J. Knell

  In his detailed bibliographic review of 1952, Robert Fay was naturally buoyant about the American contribution: “No paleontologists in the rest of the world are doing work on conodonts comparable to that of the American workers.” The Germans would doubtless agree, but perhaps for different reasons: They had been absolutely opposed to the simplistic artificiality of the American system. But the landscape was changing. Many, like Scott, who had contributed in their youth, were now senior workers, but few of these specialized in conodonts. Older workers were disappearing from the scene and with them ideas that had shaped the science. Ulrich's most important contribution took place late in life, not long before he retired from the Survey in 1932, then in his midseventies. He died in 1944, a geological colossus. His former assistant and protégé, Bassler, died in 1961 but had long been retired. Both he and Ulrich had been “eight-hour-a-day, seven-day-a-week” workers. Branson died in 1950, and although Mehl did not die until 1966, he only published one conodont paper outside their partnership. Croneis's influence on this field would also diminish as he followed a career path into university administration, eventually becoming president of Rice University in Houston and toastmaster at John F. Kennedy's last supper. He, along with many others, gained credit for bringing NASA to the city and was awarded nine honorary doctorates during his career. He also received, among many other awards, the Sidney Powers Memorial Medal of the American Association of Petroleum Geologists, the profession's highest honor.17

  What Fay could not realize in 1952 was that American control of the field was about to end. Now northern Europeans would superimpose conodont stratigraphy on the better-studied rock sequences of Germany and Sweden. It was not that America lost its place as the senior nation in this enterprise, but it was only now, with the widespread emergence of new talent, that conodont research became properly internationalized. Beckmann and Rhodes had begun to contribute to the creation of a new field, and others would follow. Looking at what they had inherited, both the chaos and the knowledge, this new generation thought the conodont was there for the taking. However, before this new wave could take control of the field there was the matter of the mess that had been handed to them. Rhodes had started to make inroads, but soon his efforts would ignite a battle of a different kind.

  Hence a chaos of false tendencies, wasted efforts, impotent conclusions, works which ought never to have been undertaken. Anyone who can introduce a little order into this chaos by establishing in any quarter a single sound rule of criticism, a single rule which clearly marks what is right as right, and what is wrong as wrong, does a good deed; and his deed is so much the better the greater the force he counteracts of learning and ability applied to thicken the chaos.

  MATTHEW ARNOLD,

  On Translating Homer (1862)

  FIVE

  Outlaws

  THE ANIMAL THAT ARRIVED IN THE 1950S, IN AMERICA AT least, had been disassembled into its component parts, cannibalized to build mythological fishes of much simpler form. These parts carried names suggesting that they were the animal, but nearly everyone knew these animals – Ulrich and Bassler's aquarium of different fishes – were mere impostors. A growing number of Americans were starting to see the fossils with Wilhelm Eichenberg's eyes – no longer as discrete things but as disaggregated skeletons demanding reconstruction. A way of seeing had changed, and with it the very idea of what might be legitimately considered an animal. Now the science needed to take control of a language that had grown absurd and return the names to the animals themselves. The proper way to do this would be to follow Eichenberg's lead and name fossils and assemblages according to the rules of zoological nomenclature. This course of action, however, was considered ill advised as it would undermine the linguistic foundations of a utilitarian science that had only just found its feet. Every word would need to be redefined, every object renamed, as terms that once referred to single things came to define sets or assemblages. A quarter century of effort would surely collapse into chaos. This, at least, was a widespread fear.

  The specter of turmoil and the politics of a close-knit field had led American conodont workers to compromise to the point of illegality. Scott's and Rhodes's invented names for fossil animals that technically already possessed names was at very least unorthodox. But now, rather than go backward and undo their earlier compromises, the conodont workers decided to stand their ground and see their actions legalized. The solution was simply to establish two parallel systems: one for naming “mere ‘nuts and bolts’”1 and one for whole things. It was not such a preposterous idea. We can speak the language of cars using makes and models – the machine's own binomial system – and yet also talk of the carburetors and sparkplugs that compose them in equivalent terms. What the conodont workers suggested was something similar: a language for discrete conodont fossils distinct from that for whole conodont animals. It was not a new idea. Carey Croneis had been developing it throughout the 1930s. It was, in fact, his personal crusade: “If any of these orphaned micropaleontologic objects…are to have correlative value (and if, paradoxically enough, they are ultimately to find parents), they must first be classed as so many different bolts, screws, and nails regardless of whether they were made in the same factory or out of the same metal.”2

  One can imagine Croneis's garage littered with pots of bolts, screws, and nails of every kind, gathered from dozens of decrepit things. By putting the screws together he gained a special knowledge of them by simple comparison; by this means and no other it was possible to become a screw connoisseur. At the University of Chicago, trays of fossil bits and pieces were similarly arranged, and with the same purpose in mind. Since the birth of micropaleontology, tiny things of all kinds had become useful to science, but their accurate identification remained problematic. Croneis's solution was to give them names of their own. Something serviceable and utilitarian – a classification of useful animal parts that would permit them to be used in correlating rocks.

  Croneis's radical solution, first published in 1938, was to remove these parts of animals from zoological nomenclature and place them in a newly constructed utilitarian classification, a military ordering (Ordo militaris) based upon the structure of the Roman army but paralleling the hierarchical ordering of systematic biology. He perhaps believed that this appeal to classical concepts added a sense of legitimacy:3

  Linnean

  Class

  Order

  Family

  Genus

  Species

  Individual

  Ordo militaris

  Exercitus

  Legio

  Cohors

  Manipulus

  Centuria

  Miles

  Croneis and his Illinois disciples now campaigned to see this change implemented. They became optimistic. “It seems…obvious that such a system must eventually be adopted,” Croneis wrote in 1941. But while Croneis attracted his supporters, few really wanted to move these objects away from the jurisdiction of the International Commission on Zoological Nomenclature (ICZN). It was the commission that established and policed the rules that ensured that each species of fossil or living animal had a unique and universally understood identity in the form of a Latinized binomial name. There was nothing more fundamental or more important in the natural sciences. Croneis was ambitious for his military scheme to be afforded the same protection, and it was to the July 1948 meeting of the once-moribund commission that two French scientists, Georges Deflandre and Marthe Deflandre Rigaud, took Croneis's scheme. In keeping with the animal that swam through minds in Chicago and Illinois (rather than Missouri or Göttingen), the plan addressed fragmentary invertebrate fossils. Any conodont workers not wishing to comply could, of course, simply claim to be students of fishes. But none of this mattered. The scheme met with complete opposition. It was rejected outright. The commission had no problem with the existence of such a plan, but only as a technical terminology outside of its control, which Croneis and his supporters knew could only result in a chaos of unc
ontrolled names.4 Their protests won a slight reprieve; the commission would permit the subject to be discussed one more time.

  So this problem of how to name and classify fossil components was “in the air” when Rhodes came to describe his conodont assemblages and in so doing redistribute the honors among his new Illinois friends. Rhodes had copied Scott's lead, but Scott had wisely referred to the elements composing his assemblages using his new adjectives. Rhodes's assemblages were composed of “form species” with Latin names of their own. Indeed, he never used the term “form species” without putting it in quotation marks, indicating that he understood its lack of status and questionable desirability. To outsiders he had created species composed of other species, and this was in a published paper. It prompted Scott Warthin, the managing editor of the Journal of Paleontology, and Eugene Eller, a scolecodont (fossil worm) stratigrapher at the Carnegie Museum of Natural History in Pittsburgh, to write to Winston Sinclair, the chairman of the Joint Committee on Zoological Nomenclature for Paleontology in America. They questioned the legality of Rhodes's actions.

  By coincidence, Sinclair was no stranger to this particular problem. A staff member at the Geological Survey of Canada, he had turned up some sponge spicules while “monkeying in things I know nothing about, like conodonts.” Wondering what he should do with them, he fell upon Scott's sponge paper and considered designating them as form genera as Scott had done. Scott was only too happy to use form names in his stratigraphic work. Finding Scott's paper in a volume “constantly being sought by spider and snake men,” he wrote to ask Scott for a copy in February 1952.5 The two men at least knew of each other before this little storm blew up.

  Sinclair's correspondence with Rhodes on the matter of form names began in the following year. It took place publicly in the pages of the Journal of Paleontology, where Rhodes had published his paper. Writing courteously, Sinclair hoped to circumvent the kind of polemics, “sarcasm and intemperate language” that often surrounded such matters, but he clearly intended to teach the Englishman “some principles which are often overlooked or forgotten.”6 Giving specific names to natural things is, of course, a scientific necessity, but it also confers a personal reward on the author by establishing him or her as an authority. Any assault on that name, therefore, is an assault on the author's integrity and intellectual offspring, which he is obliged to protect as he would his children. That this act could become so personalized was, however, beyond the ICZN’s comprehension.7 It saw only the virtuous aim of logical objectivity when in truth the names themselves resulted from complex and personal interpretations. The professional's façade could not conceal the fact that science had always been a political game.

  Sinclair made a number of points that, he argued, demonstrated the illogic of Rhodes's actions. The first was that the names Rhodes had used to describe the elements making up the assemblages referred to animals and not to things. It was certainly true that, unlike biologists studying living animals, paleontologists were much more inclined to blur this understanding. But if an individual conodont element belonged to one species, Sinclair asked, how could it, by inclusion in an assemblage, then simultaneously belong to another? Using examples from better known and more straightforward fossils, he made Rhodes's actions seem a little ridiculous.

  Sinclair and Rhodes were now engaged in a chivalrous duel, but Sinclair's initial lunge was rather ill-judged; Rhodes may have been young but he was no palaeontological backwoodsman. Sinclair sent a copy of his note to Rhodes prior to publication as a matter of courtesy. Rhodes responded immediately, and in equally diplomatic tone, but accused Sinclair of being “a little misleading.”8 This was not a problem that could be understood by extending the argument to other fossils, he said; it uniquely affected conodonts and scolecodonts, and no simple application of the rules could resolve it. The rules might result in the name Hindeodella being given to a whole assemblage, but, Rhodes argued, these and other elements were not restricted to single assemblage species or genera. Who is to say if this wouldn't be true regardless of the component chosen? In other cases, single elements in the same assemblage ranged from one form species to another, thus providing no reliable basis for naming the assemblage. The second problem arose from Sinclair's preference for Scott's use of descriptive adjectives formed from the element names. This, Rhodes felt, would produce insurmountable problems for stratigraphers and result in “the substitution of a clumsy and less satisfactory system of nomenclature.”

  Rhodes could also draw upon two decades of precedent to demonstrate that Sinclair's solution had been tried and had failed. The fact that Schmidt's 1934 application of the rules had been ignored by subsequent workers seemed to prove the point, though Rhodes made no attempt to elaborate why it had not succeeded. There were also other precedents for Rhodes's own methods, most notably Brazilian Frederico Waldermar Lange's 1947 paper on worm teeth. This paper was so “outstanding” that Lange was forced by popular demand to have it translated from Portuguese into English.9 Lange was at the time aware of the conodont problem, not least because the discovery and study of scolecodonts had happened in parallel. Indeed, Pander had unknowingly found them, and Hinde had made them a personal interest and was considered a pioneer; Stauffer was a modern master. Both fossil groups had been recognized as stratigraphically important and the fossil record of both seemed – in 1947 – to be restricted to the Paleozoic. In the case of scolecodonts this created a bizarre anomaly, for these very ancient fossils were just like modern-day forms. A huge lacuna of time separated the two in which, despite much searching, these fossils remained virtually unknown – though Lange admitted that until the eye becomes accustomed to seeing them, they do tend to remain invisible.

  Lange's paper was of interest to Rhodes because Lange had discovered some unique scolecodont assemblages (until that time only discrete or isolated fossils had been found). Now Lange could attempt a proper zoological classification of these fossil worms, but he knew he could not do so by referring to the existing taxonomy based on discrete parts. Lange was aware that Hinde had been exasperated during his own attempts to make biological species from detached components: “On one hand very different jaws are met with in the same genus; on the other, identical jaws are frequent in different genera.”10 Rhodes was clearly right to pick up on Lange's paper as it exemplified the problem he was wrestling with, but in a group of very non-enigmatic fossils. Lange's solution was to completely ignore that very imperfect classification based on what he called “formgenera” and instead use comparisons with living forms. He had independently, but like Scott and Rhodes, introduced a completely separate biological nomenclature for these assemblages, and therefore the animals themselves, stating that the “wholly artificial and transitory” system used to describe discrete jaws should be retained for stratigraphic correlation. One needed no better precedent for the actions Rhodes had taken, and Rhodes concluded, “Dr Sinclair's suggested method of nomenclature, while perhaps theoretically preferable, [is] one which cannot satisfactorily be applied in practice.”11

  The argument between Rhodes and Sinclair attracted the attention of Peter Sylvester-Bradley, who, as a member of the ICZN, doubted that Rhodes had acted illegally under the rules.12 He questioned whether the names of individual elements in an assemblage could ever be considered junior synonyms and thus disposed of once an assemblage was named after the earliest named element. The matter centered on the objectivity of the proposed relationship. Were the element species and assemblage species “objectively synonymous” (truly the same) or simply “subjectively synonymous” (merely appearing to be the same) and thus open to dispute? Since the type specimens (the actual fossil specimens described to produce a scientific name) of the element species were found separately from the assemblages, it seemed likely that the relationship would always remain subjective and beyond definite proof. It meant that Rhodes's actions were, in Sylvester-Bradley's view, legal. It was an argument grounded in the small print rather than in the intentions of the ru
les, a lawyer's reading, teasing out desirable interpretations from the inevitable ambiguities of a rule book.

  The simple issue of legality, however, was not at the heart of the problem; it was whether any dual system of nomenclature was desirable. The very purpose of having a system of rules was to ensure that any zoological species would have one name and one name only. This principle had never been overtly stated, but it had been proposed for inclusion in the rules at a recent meeting in Copenhagen. Botanists, by contrast, had been only too happy to adopt an entirely different approach. The remains of fossilized trees, for example, were hardly ever found complete. These organisms were understood first of all as discrete parts that might eventually be united into a single biological species. Consequently botanists recognized and valued form genera and form species, giving names to each type of leaf, trunk, branch, root, and seed.

  If the rules for zoological material were counterintuitive or simply unjustifiable, then Sylvester-Bradley was there to question them: “The Rules are not designed to trespass on the freedom of taxonomists to classify animals in any way they wish.” With a long history of practice and some fifteen hundred conodont species names in use, this was a taxonomic problem of considerable scale. The rarity of assemblages meant that the majority of these names would remain in use, as he felt “there seems no likelihood that it will ever be possible to assign more than a very small minority of conodonts to an assemblage.” However, that still left outstanding the conflicting interpretation of the rules by Schmidt and by Scott and Rhodes. One of these interpretations had to be outlawed, and a commission ruling would be required to decide which. Schmidt had acted legally, but his approach, it was argued, would cause grave problems for all engaged in stratigraphy. As one rising star, Maurits Lindström, remarked, “To put it drastically, it will be in the interest of micropalaeontologists to find as few conodont assemblages as possible!”13 Conodonts were in a terrible taxonomic tangle and the only solution seemed to be an appeal to the ICZN.