M. K. A. Zittel on Fossil Hexactinellida. 271 



cules, with a special dermal skeleton and lining to the stomach- 

 cavities, usually with a root-tuft. The simple square form 

 predominates in the meshes, especially of the dermal skeleton. 

 Holtenia, Craferomorpha, Bossella, Sympagella, Placodic- 



tyon^ Exiphctella, Habrodictyon^ Labaria^ Pheronemaj 



Semperella, Hyalonema. 



No one will fail to recognize the advantages possessed by 

 the grouping of the genera proposed by Marshalh The esta- 

 blishment of the family Pollakida3 especially seems to me an 

 exceedingly happy idea. It undoubtedly embraces the most 

 differentiated, multifarious, and elegant Hexactinellida, which 

 depart most widely from their precursors. Although there 

 was nothing to prevent the preservation of these forms, no 

 remains of them have hitherto been found in the fossil state. 

 The whole group appears to be confined to the present epoch ; 

 and its late appearance would consequently indicate even for 

 the Hexactinellida a progressive development from the imper- 

 fect to the more perfect. 



In other points, indeed, I am not able to adopt Marshall's 

 conceptions. If the assumption were correct that a lattice- 

 work consisting of soft sarcodic threads preceded the solid 

 Hexactinellid skeleton, the older fossil Hexactinellida must 

 necessarily have belonged, as indeed Marshall presupposes, 

 to the Synauloidas. But this is by no means the case. 

 My investigations of the fossil forms show that the con- 

 nected latticeworks without exception consist of sexradiates 

 soldered together, the axial canals of which certainly often 

 lie one over the other, and then apparently form coales- 

 cing tubes ; but in reality they are always separate, and 

 usually, as in the existing genera Farrea, Eurete, and 

 Aphrocallistes^ lie side by side in such a manner that the 

 axial threads belonging to the different sexradiates appear 

 distinctly separated. This condition of the fossil forms in- 

 duced me to make a renewed examination of the genus Sclero- 

 thaninusj for a fragment of which I am indebted to the kind- 

 ness of Dr. Marshall. The optical conditions of the material 

 at my disposal are unfortunately very unfavourable, inasmuch 

 as the axial canals only make their appearance under a certain 

 illumination, and even then very indistinctly. But after boil- 

 ing the skeleton in sulphuric or nitric acid the excessively tine 

 canals surrounded by axial cylinders appear rather more dis- 

 tinctly ; and we may then ascertain that in Scleroihamnus 

 also the latticework has been produced by the soldering 

 together of distinct sexradiates, the axial canals of which 

 meet each other and frequently lie over one another in such a 



