1858.] DAWSON LOWEK COAL-MEASURES. 73 



doids predominate, there were also large fishes of the Sauroid and 

 Coelacanth families of the Ganoids, and several Placoids. 



In the first of these families of Ganoids, the detached scales indi- 

 cate several species oi Palceoniscus* , which must have been present 

 in great numbers. 



Of the Sauroid family, there are teeth, jaws, and scales of a spe- 

 cies of Gfyrolepis ox Acrolepis, which, must have abounded at Horton; 

 and large teeth, having the characters of Centrodus (M'Coj). 



The Coelacanth family is represented by large rounded scales, 

 dentary bones, and teeth of a fish belonging, I presume, to the 

 genus Rhizodus (Owen), though the cross section of the tooth is 

 nearly round. In the cross section, these teeth present a central 

 cavity, round toward the point, elliptical further down, and sur- 

 rounded by several layers of radiating dentine, coated externally by 

 a thick layer of enamel. Their bases are deeply plicated. 



To the Placoid order belong some fragments of large spines, 

 rarely found in the Horton beds. They appear referable to the 

 genus Ctenacanthus. 



It is evident, from the presence of the remains of large predaceous 

 fishes, that these frequented the ancient Horton estuary, probably in 

 far greater numbers than the remains now found would indicate. 

 Their voracity may be one cause of the detached state in which the 

 other fish-remains occur; and accordingl}'^ many oval patches of 

 scales and teeth are associated with coprolitic matter, in a manner 

 which indicates that they formed the debris of the food of the larger 

 fishes. 



Under this head of " Pishes" I am iaclined to place certain glo- 

 bular bodies, about a line in diameter, which are very abundantly 

 scattered over the surfaces of some of the beds. They have a thin 

 carbonaceous coating, smooth externally or very finely granulated. 

 They are filled with iron-pyrites or calc-spar ; and the manner in 

 which some of them have collapsed, indicates that, when recent, they 

 were tough and membranous. This appearance and their want of 

 resemblance to any known plant of the coal-measures incline me to 

 believe that they are spawn of fishes. 



Entomostraca. — In a few layers at Horton there are quantities of 

 valves of bivalve Crustaceans of several species. Some of these have 

 ventral furrows like the " Cypridina" of De Koninck ; others are 

 ovate and smooth ; but none of them appear to be identical with the 

 Cyprids of the middle coal-measures. As the remains of these crea- 

 tures occur in but few of the fish -beds, I cannot attribute to them 

 any important part in dismembering the remains of the fishes, — an 

 office attributed to them by a recent writer. Specimens of these 

 Crustaceans have been sent to Mr. Rupert Jones, who, I trust, will 

 be able to examine and report on them. 



Annelida. — Some of the beds of sandstone resting on clay, at 

 Horton and Halfway Eiver, are completely covered with cylindrical 

 casts, apparently of the trails and burrows of worms, causing the 



* Dr. Jackson has described several species of the genus PalcBoniscus fi*om 

 Hillsborough. See ' Keport on Albert Mine:' Boston, 1851. 



