FAUNAL SUCCESSION IN THE LOWER CARBONIFEROUS LIMESTONE. 107 



succincte des Assizes du Calcaire carbonifere de la Belgique et de leurs 

 principaux facies lithographiques n ; additional details have been 

 taken from an earlier paper by the same author, ' Le Calcaire carboni- 

 fere de la Belgique.' 2 



Fossils are employed in the works above cited in two ways, viz. : — 



(a) In the recognition of levels by the maxima of particular forms or 

 groups (identified by approximate formulae), e.g., Spiriferina ' octopli- 

 cata,' Productus ' Cora.' 



This method, employed without check, is subject to large errors of 

 identification and of range: thus Spiriferina ' octoplicata ' ranges 

 throughout Z and y, and may occur at any level within this range should 

 conditions be favourable. 



(b) In the numerical comparison of a new fauna with the known 

 faunas of Tournai and Vise. 



This method (1) takes no account of the identity of genera due to 

 similarity of phase; for example, the Waulsortian facies is for this 

 reason more similar to Vise than to Tournai. (2) It does not weight 

 fossils according to their abundance ; and (3) it takes account of persis- 

 tent gentes which happen to be unrepresented at one of the localities, 

 but are present at the other, e.g., Pugnax (Rliynchonella) pugnus would 

 count as Visean, although common in the Middle Devonian. 



Within the present year two attempts have been made to compare 

 the faunal sequence in Belgium with that of the British Isles, namely : 

 (1) ' Calcaire carbonifere de Belgique : Comparaison au Sud-Ouest de 

 l'Angleterre,' by Mr. G. Delepine, 3 and (2) 'Comparison entre les 

 couches du Calcaire carbonifere de Belgique et celles de l'Angleterre,' by 

 Dr. Paul Grober, of Strassburg University. 3 



Mr. Delepine bases his partition of the series upon the recognition of 

 fossil ' bands ' (or assemblages), each of which characterises a par- 

 ticular level (under usual conditions) ; the range of a species or gens is 

 intentionally neglected. 



This method, which is essentially that involved in the use of ' marine 

 bands ' in the coal measures, leads to highly satisfactory results so long 

 as phasal conditions do not introduce a strange fauna, and so long as no 

 greater accuracy of position is sought than that involved in the large 

 zones now in use. The considerable number of forms necessary for the 

 identification of a band supply the necessary correction of errors of 

 identification, and no greater palieontological aptitude is demanded than 

 is involved in recognising a broadly formulated species. The results set 

 out by Mr. Delepine are in general accord with the subjoined correlation. 



Dr. Grober deals only with the Tournaisian, but is to be applauded 

 for his pioneer attempt to subdivide the C zone by means of variations in 

 the gentes of Caninia and Cyathophyllum. 



It is too early as yet to judge of the value of his results. I am 

 myself engaged upon the same problem in the course of a minute revision 

 of the faunal sequence at Burrington (Mendip), and Mr. A. Salee, of the 



1 Bull. Soc. beige de Giol, t. xxiii. (1909), Mem. 



2 Ann. Soc. Geol. du Nord, t. xxiii. (1895), p. 201. 

 8 Bull. Soc. beige de Giol, t. xxiv. (1910), Mem. 



