406 M. K. A. Zittel on Fossil HexactinelUda. 



appeared more definitely ; and, moreover, they could not but 

 excite the interest of the discoverer most vividly on account 

 of their marvellous beauty. Thus, in the memoirs of O. 

 Schmidt and Carter, we find the flesh-spicules very carefully, 

 and the skeleton but little considered. Greater attention was 

 bestowed upon it by W. Marshall, and especially by Bower- 

 bank, to whom we are indebted for the most thoroughgoing 

 analyses of living Hexactinellida with connected framework, 

 and for numerous figures of unsurpassable truth to nature. 



In the fossil Hexactinellida the skeleton-spicules are gene- 

 rally the only ones at all accessible to observation ; and for 

 this reason alone particular attention must be paid to them. 

 From a systematic point of view, moreover, they are by no 

 means so unserviceable as has hitherto been often supposed. 



The whole development and formation of the skeleton- 

 spicules is governed by the mode in which they unite with 

 each other. In this respect the Hexactinellida divide into two 

 natural and apparently sharply separated groups : — 



I. Lyssakina, Zittel. 



Forms in which the sheleton-sjiicules generally remain iso- 

 lated and are only united hy sarcode. 



II. DiCTYONINA, Zittel. 



Forms in which the skeleton-spicules are regularly coalescent 

 and form a connected latticework with cubical or polyhedral 

 meshes. 



The Lyssakina include the whole of Carter's Sarcohexacti- 

 nellidge, but besides these also Euplectella aspergillum and 

 cucumer. As Marshall has already shown, the cementing 

 together of the skeleton-spicules in the two last-named forms 

 is efiected by an excess of silica, which is secreted in the 

 syncytium, and fills up, at least partially, the interspaces of 

 the spicules, which are otherwise occupied by sarcode. The 

 skeleton-spicules themselves are not hampered, either in their 

 development or in their arrangement, by this secretion of 

 silica; and consequently this phenomenon, which I denominate 

 " cementation," can only be regarded as of quite secondary 

 significance. In the Lyssakina a further grouping in accord- 

 ance with the greater or less difierentiatjon of the flesh-spicules, 

 such as has been proposed by W. Marshall for his Asynau- 

 loida, would be advisable. The few known fossil rejDresenta- 

 tives of this suborder very probably possess only one form of 

 skeleton-spicules, and would therefore have to take their 



