Polytremacis and tlie Ancestry oj Helioporidm. 295 



To determine the affinities of Polytremacis we must appreciate these 

 characters : — 



B. Tlie Septal Structures. — According to Dr. Hinde [8, p. 87], 

 Neumayr [11, p. 320-1], and J. Wentzel [19, p. 490], Heliolites differs 

 essentially from Heliopora, in the possession of definite septa, which those 

 authors apparently regard as homologous with the septa of madrepora- 

 rian corals. In Heliopora there are no such septa; the structures 

 originally described as such are a series of teeth round the rim of the 

 calicular tube ; below each tooth a fluted ridge runs down the tube for 

 some distance. Neumayr, in 1899, proposed for these ridges the name 

 of " pseudosepta " [11, p. 306], and the term has been widely accepted ; 

 for, as Lindstrom remarks, the ridges are simply the projections of the 

 coenenchymal cseca.^ 



But the rule that the Heliolitids have septa and Heliopora has only 

 " pseudosepta," is not absolute. Lindstrom has figured sections across 

 Heliolites in which the septa are absent and the sections are identical 

 with those of Heliopora. For instance, Lindstrom's figure [9, Plate 1, 

 fig. 24] may be compared with a section of Heliopora cmridea figured 

 in 1895 [7, Plate 63, fig. 4]. Both sections consist of crowded poly- 

 gonal, thin-walled tubes, without any sign of septa or "pseudosepta." 

 Moreover, Nicholson has figured a calicle of Heliopora in which the 

 " pseudosepta " are more strongly developed than in the " septa " of 

 some Heliolites [12, p. 333]. 



But it is by no means certain that there is any essential difference 

 between the septa of Heliolites and the " pseudosepta " of Heliopora. 

 According to Bourne the large calicles of Heliopora are formed by the 

 fusion of nineteen coenenchymal caeca into a single cavity. The fusion 

 of the group of caeca is caused by the expansion of the central csecum, 

 which, as it grows, absorbs the adjacent parts of the surrounding caeca. 

 The outermost parts of the walls between the six peripheral caeca 

 remain for a time as radial septa ; they are finally absorbed as the 

 central cavity increases, and when it occupies the whole space of the 

 group it is bounded by a plain wall. The various stages in this pro- 

 cess may be seen along the growing edge of a lamellar corallum of 

 Heliopora. It is illustrated by a series of four figures. Fig. 6(X shows 

 a group of tubes of which the central member is slightly larger than 

 those of the surrounding series. There are no septa or septal ridges, 

 and the arrangement is identical with that of a young Heliolites in the 



* Th.e structures are here described as septa, rising the term in its descriptive 

 sense. When the septa are greatly reduced they are referred to as septal ridges, 

 analogous to the septal spines of Madreporaria. If the pejorative prefix be 

 accepted in the one case, it ought to be in the others, and Polytremacis might be 

 defined as a coral (or perhaps a pseudocoral) composed of pseudotheca, with a 

 variable number of pseudosepta, separated by pseudocoenenchyma, traversed by 

 pseudotabulse, and with a basal deposit of pseuclepitheca. 



