'£i PROCEEDINGS OF THE NATIONAL MUSEUM vol.68 



recognizable changes due to age, the many thousands covering the 

 skin of a single fish are practically alike. So far then as we can 

 learn from living fishes, the statement made on various occasions 

 that no t^o of these teeth or dermal plates are exactly alike is ut- 

 terly groundless. Moreover it is emphatically negatived by certain 

 of our collections that contain thousands of examples absolutely 

 indistinguishable in structural details except that one set is curved 

 to adapt it to the right side and another the left side precisely as 

 we found them in the supposedly related recent fishes. 



The systematic position of the conodonts has been discussed and 

 reviewed by several authors. A history of this discussion as well 

 as a bibliographic index, including illustrations of the described 

 species, forms a part of a paper by Grace B. Holmes now in course 

 of publication. As indicated above, we believe the conodonts to be 

 the teeth of primitive fishes and not necessarily all of the same 

 group. For example as shown on plate 11, illusti^ting the teeth 

 and dermal plates of certain recent fishes, Distacodus and its allies 

 seem to be related to the Myxines while the more complicated 

 Prioniodidae and Prioniodinidae show resemblance to the Selach- 

 ians. The plates described under Polygimthus and related genera 

 certainly exhibit much resemblance to the dermal plates of recent 

 sharks and should perhaps not be classified with the toothlike, true 

 conodonts. However as no true conodonts are known in Post 

 Paleozoic strata it is possible they belong to an extinct group of 

 fishes. 



In our descriptions, we have considered the typical conodonts as 

 composed of a basal portion or bar, upon which are arranged one 

 or more denticles. One of the denticles may be more developed than 

 the rest in which case it is termed the main cusp. For convenience 

 in comparison, the illustrations have been arranged so that the bar 

 is in a horizontal position. We have further considered that the 

 downwardly bent portion of the bar when this is present, repre- 

 sents the anterior end upon the assumption that this end marked 

 the attachment in the jaw. 



Our generic studies have been based upon large collections of 

 Devonian and Mississippian conodonts from many States, partic- 

 ularly from New York, Kentucky, Tennessee, Virginia, Alabama, 

 Arkansas, and Ohio. All of these collections have been studied 

 sufficiently to give a general idea of their contents, the purpose of 

 this preliminary study being to determine whether any of the species 

 from beds supposed to be of Mississippian age, contain anything 

 exactly corresponding to those from known Devonian beds. We 

 have failed to find a single instance of forms structurally identical 

 in these two periods so that the conodonts have contributed most 



