56 



THE VOYAGE OF H.M.S. CHALLENGER. 



The thick lines indicate those mesenteries which are well developed in the respective 

 genera, the thin lines those which are not so important. It will be seen that in 

 Madrepora durvillei (fig. 15) and Seriatopora (fig. 17) six mesenteries are well 

 developed, but that according to Fowler the arrangement in the one genus is precisely the 

 reverse of that in the other. In Madrepora numbers 2, 4, 6, 7, 9, and 11 represent the 

 strongly developed mesenteries, and of these numbers 4 and 9 extend lower down into the 

 ccelenteron than the other four, and are the only ones which bear reproductive organs. 

 In Seriatopora, numbers 3 and 10 are the longest. The reproductive organs were not 

 observed in this genus, but in Pocillopora, which has the same arrangement without such 

 a distinct division into three series of different lengths, apparently all the mesenteries may 

 bear reproductive organs. In Leiopathes (fig. 16) I have not ascertained which is the 

 " axial" and which the " abaxial " extremity of the stomodseum, possibly the reverse of 

 the arrangement shown, but for my present purposes it is not important. Here the 



Fig. 15. Fig. 16. Fig. 17. 



Fig. 15. — Diagram of the arrangement of mesenteries in Madrepora durvillei. Nos. 6 and 7 are the axial, Nos. 1 and 12 the 



abaxial directives. 

 Fie. 16. — Diagram of the arrangement of mesenteries in Leiopathes glaberrima. The thicker lines indicate the primary 



mesenteries. 

 Fig. 17. — Diagram of the arrangement of mesenteries in Seriatopora. Nos. 6 and 7 are the axial, Nos. 1 and 12 the abaxial 



directives. 



mesenteries numbered 1, 6, 7, 12 and 3, 10 (or 4, 9) are the primary ones. Of these 

 3 and 10, as in Seriatopora (or 4 and 9 (?) as in Madrepora), are the longest, and are 

 the only ones which bear reproductive organs. The four mesenteries numbered 3, 4, 9, 10 

 are precisely those which show an inter-relationship. One pair remains as the transverse 

 mesenteries, the others are the earliest to disappear. A very short way clown the oral cone 

 the four have become reduced to two. Thus if we suppose a combination of figures 1 5 and 

 1 7, accompanied by a replacement of the two pairs of mesenteries bordering the transverse 

 axis by one pair situated in the transverse axis, we get precisely the arrangement of 

 primary mesenteries in the Antipathidse. This does .not necessarily imply a close 

 phylogenctic relation, but is interesting as showing a similar behaviour of mesenteries 

 3. 4. 9, and 10 in Antipathidse and certain Madreporaria. 



