1865.] DAWSON COAL-FORMATIOK. 127 



It will be observed that, whether we regard the external forms or 

 the internal structures preserved, the predominant plants are Si(/il- 

 laria, Cordaites, and Ferns, with Lepidodendron and Ccdamites. The 

 substance of the coal itself, so far as its structure is preserved, 

 may be said to be principally composed of bark of Sigillaria and 

 leaves and stems of ferns and Cordaites. In regard to the propor- 

 tions in different parts of the series, little difference exists, except 

 that Cordaites and Calamites are rather more abundant in the upper 

 coals, and Lepidodendron in the lower, while the middle of the series 

 is the headquarters of Sigillaria and ferns. Eemains of aquatic 

 animals occur in connexion with a large proportion of the coals, more 

 especially in the middle of the series. This may be explained in 

 connexion with the theory of growth of the coal in situ by the follow- 

 ing considerations : — (1) It was necessary to the preservation of the 

 vegetable matter composing a bed of coal that it should be sub- 

 merged and covered with sediment. (2) On the submergence of a 

 swamp covered with standing trees and other vegetation, these would 

 prevent the passage of strong currents carrying coarse detritus, and 

 the area would be covered with fine sediment deposited in still water 

 and under conditions favourable to certain kinds of aquatic animals. 

 (3) "When the currents carrying detritus were sufficiently powerful 

 to uproot and sweep away the forests and the brakes of Calamites, they 

 would also remove or disturb the vegetable soil. It follows that we 

 should expect the more important coals to be covered with fine sedi- 

 ment containing animal as well as vegetable remains, and that beds 

 roofed with sandstone or coarse shale must either have been of small 

 area or sparsely covered with trees at the time of their submergence. 

 This accounts for the otherwise anomalous circumstance that the 

 evidences of aqueous conditions in association with the coal are pro- 

 portionally more abundant in the middle than in the upper part of 

 the Coal-measures. We may now proceed to consider the genera of 

 plants and animals separately, in their relation to the growth of coal. 



2. Coniferous Trees. — Four species of coniferous trees, referable to 

 the genus Dadoooylon, have been found in the Coal-formation of Nova 

 Scotia. They are known to me only by the microscopic structure of 

 their wood ; but on the evidence afforded by this I have named and de- 

 scribed them as new species*. One of them, D. antiquius, is closely 

 allied to D. Withami of Great Britain, and, like that species, belongs 

 to the Lower Carboniferous Coal-measures. Its structure is of 

 that character for which Brongniart proposed the generic name 

 ^^ Palceoooylon.^^ It has not yet been found at the Joggins. Another 

 species, D. Acadianum, is found abundantly at the Joggins in the 

 condition of drifted trunks imbedded in the sandstone of the lower 

 part of the Coal-formation and the upper part of the Millstone-grit 

 series. The third species, D. materiarium, is very near to D. Brand- 

 lingii of Great Britain, and may possibly be only a variety. It is 

 especially abundant in the sandstone of the Upper Coal-formation, in 



* Descriptions referred to here and in subsequent pages will be found in 

 " Synopsis of the Flora of the Carboniferous Period," Can. Nat. vol. viii., and in 

 the Appendix. 



