KONGL. SV. VET. AKADEMIENS HANDLINGAR. BAND 32. N:0 I. 19 



unknown local circumstances, thus giving origin to local varieties, but foi* the rest coin- 

 ciding in essential cliaracters Avith other specimens of the same species. 



The most commoii forms are the discoid and the hemisphcerical, more or less vaulted 

 on the superior, corallite bearing surface and with a flat basis. Near to this is the 

 lamellar colony, forming thin, incrusting or free plates, as in several of the Coccoserida;. 

 Then sphceroidal, globular, sub-globtdar and irregularly c/lobular, tuberose colonies, these 

 varieties bearing calicles on all sides, without any epitheca. A few are dendroid or ar- 

 borescent with flatly compressed or tumid branches, as Diploepora. There are also some 

 pyriform or mushroomlike colonies. A peculiar manner of growth is remarked in several 

 discoid species of Heliolites and especially amongst the Coccoseridaj : as it were, super- 

 imposed tiers growing in such a manner that a part of a colony has been arrested in 

 growth, the other mäss continuing to grow and spreading sideways covers the extinct 

 part, leaving however a free space between. 



The epitheca which is chiefly developed in discoid varieties, is a very thin, blackish, 

 often gioss}^ coating, grown in concentric rugosities and in some instances with a struc- 

 ture of its own as in Acantholithus. 



The size of the polypaina or colonies is much variable. At the largest I liave 

 found specimens measuring 30 centims. in diameter; but generally they are smaller. 

 Some nisbj have grown much larger, as may be seen from fragments. In the strata of 

 Stora Carlsö they lie in their natural position, as they once lived on the bottom of the 

 Silurian Sea, with the flat epitheca as basis. 



Propagation. It is evident that the small colonies, consisting of a primary poly- 

 pierite with beginning coenenchyma, described above, must have led their origin from 

 an ovum, which being developed into a free, swimming larva affixed itself and secreted 

 a polypary. 



The propagation through gemmation is by far more frequently observed. There are 

 in the Heliolitidfe three different kinds of it, viz. 1. Coeuenchymal gemmation. 2. In- 

 tracalicinal gemmation, and 3. Epithecal (or coenothecal) gemmation. 



1. Coenencliymal gemmation (Pl. i fig. 32, i — iv, fig. 33, pl. ii fig. 37, i — vii, 

 pl. III fig. 27 I — IV, pl. v fig. 29 I — VII.) 



The best instance of this has been found in Heliol. porosus, the procedure of 

 which, as well as of other species, is given in detail further on in the descriptions. 

 This procedure is by no means so schematically regular as Bourne has stated.^ He says 

 »that the constant number of twelve pseudosepta in Heliolites is the necessary result of 

 the formation of the calicles by the suppression of a group of seven central coenenchymal 

 tubes and the arrested growth of the adjacent walls of the twelve cells surrounding the 

 group». But even if it were so, this is not valid for the Plasmoporinee, where, and more 

 so in Propora, no reduction can take place in the coenenchymal tubuli as such do not 

 occur, at least not so pronounced as in Heliolites. In the Plasmoporinfe the twelve septa 

 are entirely new structures formed out of the coenenchymal lamellge, I suppose nearly 

 in analogy with the gemmation in Galaxea. In a certain way there is reduction, but not 



^ Heliopora coerulea p. 463 pl. 11 fig. 9. 



