572 DR. R. VON LKNDENFELD ON THE SYSTEMATIC [DeC. 21, 



at. The one I propose in this paper, which is similar to that of 

 Vosmaer, is certainly very far from being as perfect as I would like 

 to make it ; but I think that at all events it is much more likely to 

 express the relationship of Sponges in a correct manner than any 

 other existing arrangement. 



The subclass Silicea minus the order Hexactinellida comprises the 

 Sponges with a skeleton composed of tetraxial spicules (the Tetrac- 

 tinellida and Lithistidse of Marshall and other authors), the Sponges 

 with monaxial spicules (the Monactinellida of Zittel and other 

 authors), the Sponges with a horny skeleton and without spicules in 

 the supporting skeleton (the Ceraospongise or Keratosa of many 

 authors), and, finally, the Sponges without any supporting skeleton 

 at all (the Myxospongife of Hackel). 



At first sight all these forms appear connected with each other in 

 every direction by transitional forms to such an extent that it seems 

 hopeless to bring order into this chaotic mass. A careful investi- 

 gation of many forms shows that all the familiar groups Tetracti- 

 nellida, Lithistidge, Monactinellida, Ceraospongise, and Myxospongise 

 run into each other at every point, if one, however, for years 

 endeavours to find some constancy in the varying characters of any 

 chaotic mass of this kind, he at last generally arrives at an idea which 

 seems clear enough when once grasped. And then one only wonders 

 how it was that it had not been conceived a long time ago. So it 

 was also in this case. I found that all these Sponges could be very 

 naturally divided into two Orders — the above-mentioned Chondro- 

 spongise and Cornacuspongise, the first of which comprises the 

 Lithistids, Tetractinellida, and portions of the Monactinellida, together 

 with most Myxospongia; ; whilst the second contains all the Cerao- 

 spongise, and the remainder of the Monactinellids and Myxo- 

 spongise. 



We find that the ground-substance, the mesodermal intercellular 

 substance or Mesoglcea, as it is variously termed, is more or less hard 

 and cartilage-like in the Chondrospongise, and that in these the spi- 

 cules remain isolated. The spicules are either tetraxon or tylostyles, 

 less frequently styles. The monaxon spicules are monact, thereby 

 indicating their closer affinity with the tetraxon spicules. The 

 necessary toughness is given to these Sponges not by a cementing of 

 the spicules, hut by a hardening of the ground-substance. In some 

 the spicules disappear altogether, as in Oscarella, which is an aske- 

 letous form of PlaJdna, and in Chondrosia, which is an askeletous 

 sponge belonging to the Tethya group. 



In the Cornacuspongise, on the other hand, there is no tendency 

 towards a hardening of the ground-substance discernible. The 

 ground-substance remains soft and gelatinous, and the necessary 

 toughness is given to the sponge by the formation of a substance not 

 found in other Sponges, which cements the spicules together. This 

 substance is chemically and physically comparable to silk or horn, 

 and is known as Spongin. 



The spongin may become very voluminous and the spicules 

 scarce and small. They may finally disappear altogether, and then 



