MufjiliuUxii/ (ij III .\iadreporaria. 211 



ri<i;urctl the septal ]jlau of one of these young forms (I'iol. 

 15ulL, Jnne 19(J.'), ]). '.V.)). The fi<,aire shows that in this 

 partienhir specimen the details of the early septal groivtii 

 are so well preserved as to display the septa turned towards 

 one another in the regular manner revealed by serial sections, 

 but suc'.i are not indicated in Gordon's drawings ; the exo- 

 septa can also be seen in their initial relationships with the 

 entosejita, which is likewise not the case in Gordon's example. 

 Gordon himself suggests that al)sori)tion has taken place in 

 the lower j)aits of the sej)ta of his specimen, and in the case 

 of the tertiary septa {'wosepta of the present paper) acknow- 

 ledges (p. 124) that " it was impossible in all cases to tell to 

 what length they extended down into the coral." His own 

 figure and statements would prove that even the })rimary 

 septa were not all formed at the same time, though such is 

 certainly the evidence from all well-preserved material of 

 other species. Taking into account all the details which 

 Gordon offers, and studying along with them my own much 

 better series of the same species of coral, I am convinced 

 that Utile or no value can be placed upon his specimen as an 

 aid in solving the present problem. ]\Ioreover, I consider 

 that the evidence from none of my specimens could be 

 regarded as conclusive as far as concerns the number of 

 primary septa; to a certain extent the actual details would 

 for ever remain a matter of individual interpretation. 

 Assistance from decalcified silicified specimens must, in my 

 opinion, always be unsatisfactory, largely on account of 

 possible iniperfect silicification of the earliest stages. The 

 details obtainable from sections or grinding of the actual tip 

 afford by far more convincing evidence, and manifestly, from 

 the additional evidence produced in tliis paper, there is no 

 occasion for any uncertainty in the matter; there can now 

 be no reasonable doubt that the hexameral plan is that 

 characteristic of the grouj). 



To sum up, Gordon, in my opinion, (1) has failed to pro- 

 duce any evidence in favour of a ])rimary tetrameral plan in 

 the rugosids, and (2) his attempt to explain the hexameral 

 character of Loji/iop/tyUiun proliferum as resulting from 

 precocity of the first metaseptal pairs is altogether unsup- 

 ported by facts. The demonstration of six primary septa in 

 many other species is proof beyond doubt that L. i^roliferum 

 is not exceptional in its hexamerism, but conforms to the 

 rugosid type. 



Gordon is not in a position to discuss the view tliat the 

 Rugosa find their nearest modern representatives m the 

 Zoantheae. One of tiic necessary arguments for the main- 



