CARBONIFEROUS DISTRICT OF COLCHESTER AND HANTS. 255 



previously described as occurring at Hillsborough in New Brunswick ; 

 and like them they consist of dark calcareous shales abounding in 

 remains of fish. At Horton, however, the bituminous matter so 

 abundant at Hillsborough, is almost entirely wanting, and the fish- 

 scales and teeth are scattered apart, implying a less amount of vege- 

 table matter and different conditions of deposition. There are also at 

 Horton Bluff numerous bands of coarse limestone, and thick beds of 

 the white granitic sandstone already referred to, as well as gray and 

 red sandstones and marls in the lower part of the section. The most 

 interesting and abundant fossils in this section are the remains of fish, 

 which occur in incalculable numbers ; every surface in some of the 

 shales being thickly scattered over with their bright enamelled scales 

 and sharp conical teeth. Some scales are smooth, others finely 

 punctured, others marked with irregular ridges, and others with con- 

 centric lines; but all belong to the tribes of ganoids and placoids, 

 which appear to have had exclusive possession of the Carboniferous seas. 

 I have figured fragments of three of the most common species of the 

 larger fishes whose remains occur at Horton Bluff (Fig. 77). The first 

 is a species of Rhizodus^ allied to R. gracilis^ M'Coy, from the Carbon- 

 iferous shales of Gilmerton, Scotland, but differing from that species 

 in the less curved jaw, not tuberculated, but marked with irregular 

 vermicular lines, and in the thicker, finely striated, less flattened teeth. 

 Scales of this fish much larger than these figured are often seen in 

 the Horton beds. I name it R. Hard'mgi, in honour of the late Dr 

 Harding of Windsor. The jaw and scale represented at e, yj Fig. 77, 

 belong to a species whose remains are very abundant in the Horton 

 beds, and may be recognised by the pointed and deeply furrowed 

 shining ganoid scales, and the equal and flattened teeth implanted in a 

 dentary bone, whose outer surface is furrowed somewhat like that of the 

 scales. It seems to belong to the genus Acrolepis, and I have named 

 it A. Hortonensis. The spine of C/enacanthus, figured above, also appears 

 to be new. The same beds contain immense numbers of small scales, 

 probably of Palceoniscus. The appearances in these fish beds, as in 

 the bituminous limestones of the Joggius, indicate the long residence 

 of these animals in the locality, and the gradual accumulation of their 

 harder parts, as successive generations died or were devoured by their 

 larger brethren, and as the waters in which they lived wei-e gradually 

 filled up by the deposition of fine mud. We have also evidence that 

 trees grew on the neighbouring land, for trunks, branches, and leaves 

 of Lepidodcndron are very abundant, and Stigmaria Is also found. 

 In one bed, indeed, the trunks of Lej)idodendron are found rooted In 

 the erect position. They are very numerous but small, the largest 

 being only eleven inches in diameter, and their height is only six 



