Figure 13.1. A. Graptolite morphology. Drawing by Kevina Vulinec. B, C. Geniculograptus typicalis (Hall). B. Single rhabdosome, courtesy of Daniel Goldman. C. Cluster of partly parallel-oriented rhabdosomes, MUGM 29469, Cincinnatian, Butler Co., Ohio, scale in mm.
Graptolites
Graptolites are among the most distinctive fossils found in Cincinnatian strata and are also uniquely significant. Graptolites are commonly preserved in shales in a highly flattened condition, appearing like black pencil markings with a saw-toothed margin (the name graptolite in fact means “written stone”; Figure 13.1C). In some Cincinnatian limestones graptolites can be preserved in an uncompacted, three-dimensional condition. Because their skeletal structure (periderm) is organic these “inflated” graptolites can be etched free of the matrix using acid to reveal exceptional structural details (Figure 13.1B). Graptolites represent the skeletal sheath of a colonial, soft-bodied marine invertebrate whose soft parts are not preserved. Graptolite colonies existed as free-floating plankton (order Graptoloidea) or as branching, benthic colonies (order Dendroidea). The colonial skeleton (rhabdosome) housed many soft-bodied zooids each within a cup-like theca (Figure 13.1A). The walls of the thecae are constructed in a unique way that is of great importance in establishing the nature of the graptolite organism: narrow half-rings (fusellae) alternate to form a zigzag suture along the thecal tube. An outer cortical layer of collagen fibrils reinforces the fusellar layer by criss-crossing the surface; hence these are called cortical bandages. In planktic graptoloids, thecae are arranged as branches or stipes either in a single linear series (monoserial), a double series (biserial), or even triserial or quadriserial. Stipes were attached singly or in multiples by the thread-like nema to vane-like structures or possibly to gas-filled bulbs that provided flotation. Although it was once thought that graptoloids attached by means of the nema to other floating material like seaweed, it is now generally accepted that planktic graptoloids used their own means of flotation, but there is debate as to whether flotation was actively or passively maintained (Rigby and Fortey 1991).
The floating ability of graptoloids caused them to be carried great distances by ocean currents. Consequently a single graptolite species can be found over a vast geographic area, even on separate continents. This wide distribution, coupled with their relatively rapid evolutionary change over time, make graptolites some of the most ideal index fossils for biostratigraphic zonation and correlation of strata. Graptolites first appeared in the Middle Cambrian and became extinct as a group in the Pennsylvanian, but for Ordovician and Silurian strata in particular, graptolites (along with conodonts) provide the essential basis for relative age dating and correlation worldwide. The Upper Ordovician of North America is divided into four graptolite biozones on the basis of overlapping ranges of several species (Goldman and Bergström 1997).
Although several species of graptolites occur in the Cincinnatian of the Cincinnati Arch region, only a few are common (Bergström 1997). The most common and characteristic Cincinnatian graptolite is Geniculograptus (identified as Climacograptus in older literature; Figures 13.1B, C). This graptolite has a biserial rhabdosome and is very characteristic of the Kope Formation, where aggregations of stipes often occur in parallel alignment on bedding surfaces (Figure 13.1C). Two species range from the Kope through Fairview Formations (Bergström 1996a). Two biserial graptolites occur in the Arnheim Formation (Richmondian): Orthograptus quadrimucronatus and Arnheimograptus anacanthus (see Bergström 1996a). Several species of Mastigograptus, a delicate, bush-like dendroid graptolite, occur in the Kope, Arnheim, and Waynesville Formations.
The zoological affinities of graptolites were for many years among the most challenging problems in paleontology. Graptolites were classified with several different groups, including cephalopods, cnidarians, bryozoans, and hemichordates, or were considered to be unrelated to any living group (Bulman 1970). Research by the Polish paleontologist Roman Kozlowski (1966) noted several similarities between graptolites and the living hemichordates called pterobranchs that argue strongly for a close evolutionary relationship. Pterobranchs are a group of small, marine, tube-dwelling invertebrates that are classed together with the acorn worms in the Phylum Hemichordata on the basis of embryological similarities; there are only three living genera of pterobranchs (Barnes 1987). The creeping tube constructed by the living pterobranch Rhabdopleura has the unique fusellar half-ring and cortical structure found in graptolites (Bulman 1970). In addition, some encrusting types of graptolites have a black stolon preserved within the tubes that is very similar to the structure of the stolon in pterobranchs that interconnects zooids within the colony. The identification of the protein collagen in the tubes of both graptolites and pterobranchs indicates not only their close relationship, but also affinity to chordates, in which collagen forms the connective tissue (Armstrong et al. 1984). The extinct graptolites are thus generally treated as an extinct class within the Hemichordata. Although the soft part structure of the graptolite zooids remains unknown, hypothetical reconstructions suggest that the zooids probably had paired tentacle-bearing arms that were extended for purposes of suspension feeding and deposition of the outer cortical bandages of the thecal wall.
Conodonts
It is ironic that perhaps the most significant Cincinnatian fossils, in a geological sense, are also among the least conspicuous to most observers. These are the conodonts, tooth-like microfossils (< 1 mm), so unlike any other fossil or living organisms that they were regarded until recently as representing a distinct phylum. Conodonts can be found throughout the Cincinnatian by dissolving blocks of limestone in acetic acid, although they can also be found in disaggregated shales. Because conodonts have a calcium phosphate composition, they are insoluble in acetic acid. Conodonts are conspicuous in the acid-insoluble residue scanned under a dissecting microscope because of their unique forms and a beautiful amber color (Plate 5B).
Figure 13.2. Cincinnatian conodonts. Those shown are among the most common forms. All are from the lower Richmondian Stage near Brookville, Franklin Co., Indiana. A, G, H. Plectodina tenuis (Branson and Mehl). A. Pb element. G. M element. H. Sc element. B. Oulodus oregonia (Branson, Mehl, and Branson), Pb element. C, D. Amorphognathus ordovicicus Branson and Mehl. C. Sc element. D. Pa element. E. Drepanoistodus suberectus (Branson and Mehl). F. Phragmodus undatus Branson and Mehl, S element. I. Rhodesognathus elegans (Rhodes), Pb element. J. Aphelognathus grandis Branson, Mehl and Branson, Pb element. Scanning electron micrographs courtesy of Stig M. Bergström, approximately × 130. P elements are principal elements, a and b denoting position in the apparatus, M elements are medial elements, and S elements are symmetry series elements, c denoting position in the apparatus.
Conodonts have a wide range of shapes, including single cones, multi-pronged “teeth,” serrated blade-like shapes, and so-called platform-type forms (Figure 13.2). Most conodonts have a tooth-like appearance, leading to the assumption that they were used as teeth. However, conodonts also show regeneration. This suggests that at least some conodonts were embedded in soft tissue by which they were secreted. Individual conodonts, called elements, were given species names by earlier workers. However, the discovery of rarely preserved assemblages of different elements in rock matrix or fused together led to the recognition that elements were arranged in bilaterally symmetrical assemblages of pairs of elements, each of which is called an apparatus. Although few apparatuses are preserved intact, it has been possible to determine which elements found in a sample likely formed an apparatus on the basis of statistical analysis of the ratios of commonly associated elements. Modern taxonomists of conodonts attempt to include the six or seven different elements of a given apparatus under a single species name.
Because the tooth-like elements apparently were the only mineralized parts of the organism, the identity of the conodont-bearing organism and the nature of its soft part anatomy have been among the great mysteries of paleontology. A major breakthrough came in 1983 when a fossil of an eel-like, soft-bodied organism from the Lower Carboniferous of Scotland was shown to have a conodont apparatus at one end and fin-like structures at the opposite (Briggs et al. 1983). This “conodont animal” is only 4 cm long and shows little of its internal anatomy, save for a midline suggesting bilateral symmetry. Although it bears similarities to both chaetognaths and chordates, the authors concluded that the conodont animal belongs to a distinct phylum. More recent phylogenetic studies based on more extensive data place the conodonts within the chordates, close to the living lampreys (Donoghue et al. 2000). Because conodont animals had mineralized calcium phosphate elements, among other characteristics, they are considered to be vertebrates, even though they lacked internal skeletons. The conodont animal could be considered to have been the fish of the Cincinnatian sea. However, given their small size and great differences from anything like present-day bony fish, the title of this book, A Sea without Fish, remains valid. The foregoing introduction is largely based on Clark (1987). For more information about the morphology and affinities of conodonts the reader should consult Aldridge et al. (1993), Aldridge and Purnell (1996), and Donoghue et al. (2000).
Conodonts of particular types are found in marine rocks over a very wide geographic range, even intercontinental in extent. This wide distribution suggested that the conodont organism was pelagic long before the fin-bearing conodont animal was discovered (Clark 1987). Conodonts apparently lived as swimming members of the marine nekton ranging from close to the sea floor to various positions in the water column. The tooth-like apparatus might have been used to seize or strain food items from the water.
In addition to their wide geographic range, genera of conodonts are restricted in their vertical (stratigraphic) ranges throughout their Middle Cambrian through Triassic geologic record. These attributes, together with abundance in marine strata as microfossils, make conodonts exceptionally useful for biostratigraphic subdivision of the Paleozoic sedimentary rock record. For the entire Upper Ordovician (above the base of the Lexington Limestone) in the Cincinnati region, Sweet (1979) recorded thirty-five conodont species representing twenty genera (Figure 13.2). Most genera occur also in Europe, Asia, Australia, or Africa, but species are more geographically restricted. Most Cincinnatian conodonts represent the North American Midcontinent Province, but some are components of the North Atlantic Province that is also represented in Europe. The type-Cincinnatian section spans all or parts of six conodont-defined biostratigraphic zones (Webby, Cooper, Bergström, and Paris 2004). Although many conodont taxa have long stratigraphic ranges within the Cincinnatian, variations in relative abundance are probably related to differing depth preferences among species (Sweet 1979, 1996).
Figure 13.3. Reconstruction of the Ordovician jawless fish, Astraspis desiderata Walcott, from the Harding Sandstone of Colorado. Length about 13 cm. From Cowen (2005, 86, figure 7.4). Reprinted with permission of Blackwell Publishing.
Ordovician “fish”
Conodonts may have been the only representatives of the vertebrates in the Cincinnatian sea, but fossil evidence from many localities elsewhere shows that some of the earliest fish did indeed exist during the Ordovician Period. The Ordovician Harding Sandstone of Colorado contains abundant, tiny plates composed of an enamel-coated dentine, named Astraspis desiderata by Walcott in 1892. In 1997 Sansom and others illustrated an extraordinary fossil from the Harding with a compressed, nearly complete, fish-like body that proved the long-held assumption that the tiny plates formed a head shield (Figure 13.3). Eyes and a series of openings probably related to respiration are preserved. A similar fossil fish from Ordovician rocks in Bolivia shows that these early fish had blunt, rounded heads, an elongated fish-like shape with a tail fin, but lacked bony jaws and separate fins. These jawless fish are called agnathans, but other Ordovician fossils represent the earliest jawed fish or gnathostomes (Sansom et al. 2001). Thus, a variety of early fish had already evolved by Cincinnatian time, but they are unknown from the Cincinnati region. Had these early fish been present in the Cincinnatian sea, it would seem reasonable to expect their mineralized plates to be preserved in the limestones or shales. Although their absence may indicate that they preferred an environment not represented in the type-Cincinnatian, the potential for their eventual discovery should not be overlooked.
