157 



In this species we find within the same colony some zocecia, wliich are provided 

 with a frontal shield similar to that in Membraniporella nitida, and some which 

 have either only 2 — 4 short distal spines or besides these a varying number of 

 longer, unconnected ones. A partial coalescence of s|)ines may also occur in several 

 species. Thus, I find the first pair of spines coalesced in no small number of 

 zod'cia on a colony of Membraniporiiui pijrula, Hincks from Victoria, and in the 

 Meinbr. defensa described by Kirkpatrick a number of opposite spines (in the 

 figured specimen 4 pair) may sometimes be connected. Such a coalescence of 

 o[)posite spines also takes place in the whole length of the frontal area in Sto- 

 lonella clausa Hincks, which belongs to the Bicellariidae, and a frontal shield formed 

 by 5 coalescent broad spines occurs in Petalostegus hu-oriiis, which I have thought 

 it most correct to refer to the same family. 



The natural consequence of the view expressed above would then be the 

 splitting up of the family Cribrilinidae and the grouping of its forms with such 

 forms of the Membr(tniporidae. to which they are most closely allied. As however 

 the latter family cannot be regarded as natural either, it must perhaps be broken 

 up into a larger or smaller number of smaller families, and the forms, which 

 have a frontal shield, must be divided among them. My material of both these 

 families is however too small for me to venture upon making definite proposals 

 for a final arrangement of all these forms, and therefore for the present I prefer 

 to keep the two families unaltered. As the members of this family have their 

 nearest relations in the family Membraniporidae. it is quite natural, that we should 

 find' a similar extent of variation in most structures. The cryptocyst forms however 

 an exception in this respect, as it is either completely wanting or appears only 

 as an extremely faint margin within the spines (MembraniporelUt). This is a natu- 

 ral consequence of the presence of the frontal shield, the latter making such pro- 

 tection unnecessary as in uncovered Membraniporidae may be rendered by the 

 cryptocyst. The frontal shield may be of very difVcrent extent, occupying at times 

 the entire frontal surface and in other cases but a smaller part of the latter. It 

 is much reduced in certain forms occurring in the Danish chalk-formation. While 

 in some cases we find an opercular valve only, a wholly chitinized (simple or 

 complex) operculum is often present as in the members of the family Scriipii- 

 cellariidae, in which the opercular si)ine is so strongly develo|)ed that it forms 

 part of the boundary of the operculum. As in Membraniporidae, the rosette-plates 

 may be multiporous, uniporous (e. g. Membraniporella dislansj or there may be 

 pore-chambers. The heterozooecia may appear both as independent and dependent 

 ones, as avicularia and as vibracula, and rather frequently we find a calcified 

 transverse bar between the opercular and the subopercular area. The ooecia are 



