M. K. A. Zittel on Fossil Hexactinellida. 423 



the Eocene sandstone of Brussels*, and from Miocene sands 

 at Ruditz, in Moravia, and of a Miocene Aphrocallistes from 

 Russia, I know of no Tertiarj Hexactinellida from Northern 

 and Central Europe. 



But even in the South-European Nummulitic formations, 

 to which, at least in part, a deep-water origin is ascribed, it 

 is remarkable that they occur only as great rarities. The 

 only certain evidence of them consists in an Eocene Guet- 

 tardia, which D' Archiac has described from the neighbom-hood 

 of Biarritz. 



This hitherto unexplained gap, however, is partially filled 

 by A. Pomel's important discovery of numerous Miocene 

 sponges in the province of Oran. Among the North-African 

 Hexactinellida the genus Craticularia [Laoccetis, Pom.) takes 

 the first place by its astonishing abundance of forms : species 

 of Aphrocallistes [Badinskia, Pom.), Tretostamnia^ Pom., 

 and Placochlcenia^ Pom., as well as a considerable number of 

 Lithistida, are also described. 



If, therefore, the fossil Hexactinellida, by their peculiar 

 geological distribution, prove to be deep-sea dwellers almost 

 as clearly as their living relations, we obtain in these orga- 

 nisms an important datum for judging of the mode of forma- 

 tion of geological deposits. 



The limitation of the fossil Hexactinellida to deep-sea 

 deposits, however, also necessarily involves their intermittent 

 occurrence, separated by long interruptions. In formations 

 which are at present known only under a littoral facies, there 

 are no Hexactinellida. The different sponge-horizons are 

 therefore also in part separated by enormous intervals of time. 

 Thus, for example, the Silurian forms are immediately fol- 

 lowed by the Upper Jurassic (no true spongitic deposits are 

 known in the Devonian, Carboniferous, and Dyas) ; and these 

 agahi are separated by a wide interval from the Middle and 

 Upper Cretaceous. This best explains the fundamental dif- 

 ferences in the successive sponge-faunas in the Silurian, the 

 Jura, the Cretaceous, and the Miocene. Under these cir- 

 cumstances we ought rather to wonder that there are any 

 genera common to two formations, than that the Jurassic and 

 Cretaceous Hexactinellida, for example, present great differ- 

 ences. 



Perhaps there are few divisions of the animal kingdom, 

 capable of preservation, of the phylogeny of which palaeonto- 

 logy furnishes so fragmentary a picture. Our entire know- 

 ledge of the fossil Hexactinellida is limited to isolated and 



* Rutot,/.r. pi. iii. figs. 3.3, 34. 



