606 J. E. MARE ON THE PEEBEVONIAN 



mixture of other remains, is superposed to a band containing animals 

 which belong to his Third Fauna, or Upper Silurian;" and a few 

 lines further on, " these alternations of life indisputably connect the 

 Lower and Upper Silurian rocks in one system, through an inter- 

 change of a considerable number of their respective fossils." It is 

 needless to describe the theory of M. Barrande : this theory has 

 been attacked by various authors. In 1852 Lipoid tried to account 

 for the phenomena in Bohemia by foldings, but afterwards aban- 

 doned this explanation. Prof. Krejci afterwards supposed them to 

 be due to faults; both he and Lipoid, in letters inserted in M. Bar- 

 rande's ' Defense des Colonies ' (iii. pp. 80, 82), declared that their 

 former ideas on the subject were not correct. In 1868, D'Archiac, 

 from a consideration of certain phenomena in the Carboniferous 

 rocks of Belgium, which had been accounted for by M. Barrande's 

 theory, but afterwards proved to be due to physical disturbances, 

 threw doubt on the theory as applied to the Bohemian phenomena. 



The fauna occurring in the colonies of Bohemia was supposed to 

 have existed in the north of Europe ; but this view, as shown by Lin- 

 narsson (c/. abstract in Geol. Mag. dec. ii. vol. v. p. 282), can no longer 

 be held. In fact it has been shown by Lapworth, Linnarsson, &c. 

 that the Birkhill Shales and their Scandinavian equivalents, which 

 contain a corresponding fauna to that of the colonies, are not to be 

 referred, as was formerly supposed, to the Bala group, but are much 

 newer. Other fossils than Graptolites occurring in the colonies 

 have also been found in Britain, and a table of these is given by 

 Barrande (Defense des Col. iv. p. 131), in which he argues that they 

 all occur in equivalents of his Second Fauna in Britain, that is, in 

 the Upper Cambrian. In analyzing this list, we find the following 

 recorded as Bala : — Cheirurus insignis, Beyr., C. himueronatus, 

 Murch. ; these do occur at that horizon. Splicer exochus mirus ; all 

 the specimens from the Bala of Britain referred to this species, 

 which have been examined by Tornqvist, are referred by him to 

 Sphcereccochus angustifrons, Ang. (CEfvers. af K. Yetenskaps-Akad. 

 Forhandl. 1879, Ko. 2, p. 70). Gardiola interrupta, from the Bala 

 of Coniston ; this was when Profs. Harkness and Nicholson consi- 

 dered the Coniston Flags to be Bala. Atrypa reticularis and Stro- 

 phomena euglypha are from the Llandovery Rocks. Hence the fauna 

 of the Colonies is by no means to be found in the Upper Cambrian 

 rocks of Northern Europe ; on the contrary, it occurs at the base of 

 the Silurian, and therefore it cannot have migrated from the N.W. 

 In fact there is no locality known where this fauna is not charac- 

 teristic of the base of the Silurian. The same forms of Graptolites 

 occur in Sardinia, as figured by Meneghini ; but their horizon is 

 unfortunately unknown. 



Nobody now denies the occurrence of migrations, so ably treated 

 of by M. Barrande in the third part of his ' Defense des Colonies,' 

 in which species occur again in beds of the same great group later 

 than those in which they are first found, if the conditions be 

 favourable for their return, but usually mixed with other forms. 

 This, however, is very different from what is required by the theory 



