MORPHOLOGY OF THE MADREPORARIA. 



and situated at equal distances apart, thereby presenting a more 

 truly radial disposition. On the other hand in the mature coral- 

 lite of Hadrophyllum the septa of the alar region do not reach 

 the radiate condition ; they retain their unipinnate arrangement 

 throughout, and, as a consequence, the alar regions remain 

 sharply separated from the rest of the corallite both by a special 

 arrangement of the septa and by an interval. 



The series of sections of Strcptclasma further proves that the 



alar fossula is situated on the 



-f 



dorsal side of the alar septum ; 

 the latter takes no direct part 

 in forming the depression, but 

 merely constitutes its ventral 

 boundary. As shown in Fig. 

 2, the right and left alar septa 



are the ventro-lateral pair of 

 the six primary septa, and all 

 the additions in the middle in- 

 terseptal chambers necessarily 

 take place dorsal to them. 



FlG. 4. A stage a little beyond that in 

 In sections the new septa are pjg 3 A corresponding pair of septa () 



inclined by their inner border has appeared within the two ventro-lateral 



chambers or principal quadrants, each 

 member inclined towards a ventro-lateral 

 (alar) primary septum. 



towards the dorso-lateral pair 

 of primary septa, and are thus 

 more nearly parallel with the 

 ventro-lateral (alar) septa. 1 Sometimes, as in Anisophylhtm agas- 



1 In giving the septal plan of a tetracorallid as viewed from above Delage & 

 Herouard ( " Traite de Zoologie Concrete," II 2 , p. 692, Fig. 973) project the newer 

 septa of the principal quadrants upon the cardinal septum and parallel with the alar 

 septa, while those in each counter quadrant are projected upon the alar septum and 

 are parallel with the dorsal directive or antipodal septum. The figure of Hadrophyl- 

 lum here given, and also those showing the septal development of Streptelasina rec- 

 tum, prove that this is not the true relationship of the inner ends of the metasepta 

 within the calice. In each quadrant the inner end of each septum is from its origin 

 directed towards the next older member of its own series. I have met with no in- 

 stance in which the later developed septa are inturned towards and fused with the 

 alar and axial directive septa, as seems to be usually assumed from the appearances 

 presented by the external ridges and grooves alone. The explanation which Bourne 

 gives (Lankester's "Treatise on Zoology," Art. Anthozoa, p. 73, Fig. 35) of the 

 schematic representation of the septa of a zaphrentoid coral is manifestly self-contra- 

 dictory as regards the sequence. 



