The ?<latural History of the Frilled Shar\ 307 



gihhosus Ag. (our Text-figure 25) with those of Chlaynydoselachus. In Diplodus the 

 two lateral cusps are large and blunt, the central one almost rudimentary, as may be seen 

 in Text-figure 25 A. Agassiz, ^ shows another tooth of this extinct shark (our Text-figure 

 25B). Here the two outer cusps somewhat approach the shape of the teeth of Chlamy 

 doselachus, but between them is a rudimentary structure called a "button" by Cope. 

 Let us contrast these teeth with one from Chlamydoselachus (Text-figure 10). Here we 

 have a central long rather slender cusp, flexible even in the fossil form (Lawley, 1876), 

 and at its base are the two rudimentary cusps, while outside of these are the two 

 lateral cusps which are almost as large as the central one. We cannot see any essential 

 likeness. 



Gill at first was inclined to acquiesce in Cope's views of the relations of Djd)i7nodus 

 or Diplodus, but in his next communication (April 11) he brought his unrivalled knowledge 

 of the literature of fishes to bear on the question. He traced (1884.2, p. 429) the history 

 of these names, noted that Cope in 1883 had substituted the former name for the latter, 

 and concluded with the statement that 'The distinguished naturalist [Cope] was 

 evidently not acquainted with the researches of his predecessors." Furthermore Gill 

 was able to show that the names Diplodus and Didymodus are synonymous with Pleura- 

 canthus, or with the names of other related forms of the family Pleuracanthidae, all 

 members of which had prominent dorsal spines. Gill was of the opinion that 

 Didymodus not only has neither generic nor family relationships with Chlamydoselachus 

 but that it belongs to a different order. Lastly he ventured the prediction that a 

 study of the anatomy of Chlamydoselachus would show structures most like those of the 

 Notidanidae, but probably more primitive — a relationship now generally accepted. 



Following the publication of Gill's note. Cope studied the literature, and acknowl- 

 edged (May 30) that "On some points I stand corrected." Recognizing the fact that 

 Diplodus had a dorsal spine, he agreed (1884.3, p. 645) that Chlaynydoselachus must be 

 separated from it on account of the structure of the dorsal fin, but he queried "whether 

 such a spine is concealed in Chlamydoselachus or not." To this the answer is that there 

 is none. Furthermore, it may be remarked that the Pleuracanthids seemingly lack a well- 

 developed dorsal fin, and that the spine referred to is attached to the dorso-posterior 

 region of the skull. 



However, on March 7, 1884, Cope had read before the American Philosophical 

 Society a paper on the skull of Didymodus, based on the study of twelve crania from the 

 Permian of Texas. In his paper published in the Proceedings (July 1), he again affirmed 

 (1884.4, p. 588) that the teeth of this shark do not differ generically from those of Chlamy- 

 doselachus, but that the fishes must be separated on the differences in structure of the 

 dorsal fins. His text does not indicate that he had ever seen the teeth of Chlamydoselachus, 

 and his figures of the teeth of Didymodus certainly show no generic resemblance to those of 

 Chlamydoselachus. This will be recognized when one compares his figure of a tooth of 



^Agassiz, L. Recherches sur les Poissons Fossiles. 1843, Atlas, Tome III, plate 22b, figs. 1 and 4 {Diplodus gibbosus), fig. 11 

 {Cladodus mirabilis), and fig. 21 {Cladodus acutus). 



