No, 2.] THE "colonies" of m. barrande. 201 



In the first place, it seems certain that during the existence of 

 the last phases of the second fauna in Bohemia, the first phases 

 of the third fauna had become more or less fully developed in 

 some other country hitherto unknown. 



Startino; from this centre of diffusion, migrations must have 

 taken place at different epochs into Bohemia, during the whole 

 of the deposition of the thick band d 5. 



On every occasion these mi<>;rations must have jiiven rise to 

 colonies, which are placed on the same horizon, and consist of 

 graptolitic schists, almost always accompanied by flows of trap, 

 and often containing; calcareous concretions. 



In consequence of inauspicious conditions, and from the cessa- 

 tion of these schistose and calcareous deposits, all the colonies 

 must have enjoyed a relatively short existence during the period 

 that the Bohemian area was occupied by the second fauna. 



The appearance of the colonies coinciding constantly with the 

 graptolitic deposits, we are compelled to attribute both equally to 

 the influence of currents arising in the same quarter. 



The introduction of intermittent currents into the isolated 

 basin of Bohemia seems to have been caused by oscillations of 

 the land, connected with the production of the traps which occur 

 so frequently in bands d 5 and e 1. 



In all cases, the colonial species appeared on difibrent horizons 

 without being able to establish themselves permanently in Bohe- 

 mia during the last phase of the second fauna. 



After the complete extinction of the second fauna, however, 

 and after a prolonged intermission, during which the Bohemian 

 basin appears to have been deserted, a new immigration, arising 

 from the same foreign centre, must have invaded the Bohemian 

 sea, and must have succeeded in permanently establishing itself 

 there. (I may remark here that few palaeontologists would admit 

 that the presence of a considerable mass of unfossiliferous beds in 

 the midst of a fossiliferous series, necessarily implies a period in 

 which life did not exist, as above assumed by M. Barrande. 

 More probably the local conditions were such as to cause a local 

 migration of the existent fauna, or such as not to allow of their 

 preservation in a fossil condition. There certainly do not seem 

 to be sufficient grounds for the assumption that the whole of the 

 second fauna of Bohemia died out during the deposition of the 

 upper part of d 5, and the absence of fossils might be partially 

 accounted for by the lithological nature of the deposits in ques- 



