FOSSIL FISHES. 43 



by Owe. I and Morse, be the teeth of mollusks, for tliey strongly- r>3semble 

 them in their peculiar and varied forms and their chitonous composition, 

 bnt if the teeth of mollusks, these must ha^e been shelless, for no mol- 

 luscous shells have been found in the formation which contains them, 

 and it would be somewhat singular if the Mollusoa were represented in 

 the sea from which the Huron shale was deposited, only by naked species. 



3. In the first volume of this Eeport I ventured to suggest tliat the 

 Conodonts might be the dermal ossicles of ancient fishes. Among the 

 Elasmobranchs there are some of which the external surface is protected 

 by a shagreen composed of divergent, acute ossicles, not very unlike 

 some of the Conodonts in form and composition. We know of no sha- 

 green, however, composed of such neat, regular bodies as the Conodonts 

 are, and the conjecture that they may be the shagreen of sharks is as yet 

 supported by little evidence. 



I also called attention to the fact that tlie scutes of the Sturgeon, about 

 and within the brancliial apertures, have almost exactly the outline of the 

 simpler Conodonts, but further observation has led me to consider this 

 resemblance as accidental, and without zoological significance. 



4. Waiting further evidence of the nature of these interesting organs, I 

 take the liberty of offering, as a possible and plausible explanation of the 

 enigma, the theory that they are the teeth of Cyclostomous fishes. If 

 any one will take the trouble to compare the Conodonts with the teeth of 

 Myxitie and Bdellostoma, he will find a very close and remarkable sim- 

 ilarity between them. Indeed, except that they are very much larger, 

 the teeth of Bdellostoma polytrema are almost exactly like the Cono- 

 donts represented in Figs. .12, 14, and 16. The teeth of Myxine gluti- 

 nosa are even more like these in size and delicacy of structure, and 

 scarcely less so in form, than those of Bdellostoma. In composition and 

 internal stru.cture, the teeth of the modern Marsipobranchs seems to be 

 almost identical with the Conodonts, and the resemblances which they 

 present are throughout so strong that few will make a comparison between 

 them without being convinced that they are nearly related zoologically. 

 From the low place held by the MarsipolranoMi in the zoological scale 

 we might naturally suppose that they existed in considerable numbers in 

 the Palaeozoic seas, and it has been a matter of some surprise that no 

 traces of them have been heretofore recognized in any ancient strata. 

 This fact has been explained by the suggestion that none of the Cyclo- 

 stomes have organs composed of tissues that would resist decay. 



This explanation is, however, not altogether satisfactory. It is true that 

 nothing but the teeth would be likely to be preserved, but even if com- 

 posed of horn-like tissue, as in Petrorayzon, they should have left some 

 traces when buried in the finer sediments ; wlien harder, like those of 



