77 



for the lelationship. On the other hand, the mode of calcification seems to have 

 only a slight systematic importance, hecause it often varies, not only within the 

 genus Init also within the species (see pai;. fj). 



The frontal wall. As we have already fully discussed the ditTerences, which 

 occur ill llu' slrucliire of the fronlal wall under the morphology of the zoa-cium, 

 we may content ourselves hei'c with a brief reference to this (piestion. While 

 Jul lien and later Canu attach such a great systematic importance to the crypto- 

 cyst, that they on the basis of it divide the Cheilostomata into two main divi- 

 sions: Diplodenudta and Monothtnuthi, we can only from the investigations we 

 have made on its extension in the difi'erent families (see pag. 13 — 16) regard it 

 as an auxiliary character. It is for the present not easy to judge to what extent 

 it can be used as such, since in many cases it is difficult to decide, whether 

 dried or even spirit Bnjozini have a cryptocyst or not. We may for instance 

 mention, that out of the whole material, which the zoological Museum of Copen- 

 hagen jiossesses of Esclutroides coccinca, only a very few colonies of the genus 

 show a distinct covering membrane. \\'hile the frontal wall in some forms, e. g. 

 in the species of the genus Electra, the members of the families Hippolhoidae 

 and Catenariidae is only formed of a gymnocyst, and in others e. g. the Onijcho- 

 cella species only of a cryptocyst, both modes of calcification are represented on 

 this wall in most of the Cheilostomata. The cryptocyst shows a veiy remarkable 

 condition in Steganoporetla and Thalamoporella, as it descends more or less deeply 

 into the zofficium through two (more seldom one single) openings proximally to 

 the aperture and may even reach the opposite wall. Still more peculiar is the 

 condition it shows in the genera Cellularia and Membranicellaria, because the 

 surface of the colony is divided by a network of ridges into a number of areas, 

 which do not correspond at all with the real zott'cia. We have also seen that 

 the calcification of the frontal wall may take place to a very varying extent, so 

 that we can find all possible transitions between a cpute uncalcified and a (|uite 

 calcified fronlal wall, and the characters derived from the dillerent extension of 

 the calcification are therefore more or less relative. 



The pores. I have already called attention to the fact, that a number of the 

 so-called pores seem to have a similar structure to that of the rosette-plates; but 

 as time and material have not allowed me to carry through such an examina- 

 tion everywhere, I shall here under the name of pores include all [)ore-like 

 foruialions, which in contrast to the rosette-plates are situated on free surfaces. 

 The pores are certainly amongst the formations, which have the least systematic 

 importance, because in most families and genera where they ajjpear, they are 



