Introduction
IN RECENT YEARS there have been widely scattered reports of fossils believed to be cephalopod aptychi from the Paleozoic of North America (Brady, 1955, but see Yochelson, 1971; Closs, et al., 1964; Saunders and Spinosa, 1974; Saunders and Richardson, 1979; Thompson, et al., 1980; Yochelson, 1983; Kues, 1983; Mapes, 1987; Landman and Davis, 1988; Harper, 1989). Although several of these reports claim to be one of only a handful of such findings, many more specimens are available for study. The collections of the Cleveland Museum of Natural History contain over a hundred such specimens from the Late Devonian Chagrin and Cleveland shales.
These fossils are preserved as flat, glossy black, carbonaceous smears in the dark shales. They are marked with fine concentric ridges which parallel the margin or are truncated by it along the periphery. The ridges resemble those sometimes seen in brachiopods or bivalves, and the general outline could be suggestive of some early crustaceans. Similar fossils also have been identified as fish scales, barnacle plates, or perhaps gastropod opercula.
The earliest description of these Devonian fossils in North America identified them as a form of phyllocarid crustacean, Spathiocaris (Clarke, 1882). Woodward (1885b), however, acknowledged that some "phyllocarids" could, in fact, be goniatite aptychi. Later, Clarke (1902) expressed his doubts in the matter, admitting that they could be cephalopod aptychi or brachiopod fragments instead. This study was begun in an attempt to determine their affinities.
It was expected that the application of new methods might yield additional insight into the subject. Examination of the microstructure of these structures using the scanning electron microscope might reveal details of their formation and growth which would permit a more definitive identification. Determination of the chemical composition of these structures by x-ray spectroscopy while they were in the electron microscope might also confirm their affinities. Cephalopod shell material is aragonitic, whereas their mandibles are calcitic (Lowenstam, et al., 1984). Aragonite frequently contains strontium as a significant trace element. Inarticulate brachiopod shells and arthropod carapaces are based on calcium phosphate, not calcium carbonate. New chemical data would not necessarily be definitive, however, for bivalves are predominately aragonitic.
The principal purposes of this study are to examine and describe the "spathiocarids" of the Cleveland and Chagrin shales, investigate their relationships with similar Devonian taxa, and attempt to provide solid identification of their nature, if possible. This latter goal was not fully realized, but two of the most likely alternatives have been eliminated. It is probable that these fossils should be referred to the Cephalopoda.