58 THE VOYAGE OF H.M.S. CHALLENGER. 



If, however, the researches of Lacaze Duthiers should be confirmed, it appears to me that 

 there is an important connection between the order in which the first twelve mesenteries 

 are developed in Actinia, Heliactis, &c, and their relative importance in Antipatharia, 

 a connection which may probably throw light on the phylogenetic relations of the two 

 orders. If I have understood Lacaze Duthiers aright, the first twelve mesenteries in 

 Actinia, &c, are developed in an order which may be explained by a reference to fig. 

 16. The first to be developed are numbers 3 and 10 of that figure, the second, numbers 

 6 and 7, the third, numbers 1 and 12, the fourth, numbers 5 and 8, the fifth (?), numbers 

 2 and 11, the sixth (?), numbers 4 and 9. Lacaze Duthiers does not number the mesen- 

 teries in the same order in his figures of Actinia as in those of Heliactis (Sagartia). 

 Now if the order which I have indicated should prove to be the correct one, it precisely 

 corresponds with the relative development of the mesenteries in Leiopathes and other 

 Antipathic! ae. Numbers 3 and 10 are in Leiopathes the longest, numbers 1, 6, 7, 12 next; 

 numbers 2, 5, 8, and 11 come next; and finally, numbers 4 and 9 are the shortest of all. 

 It is to be noted further that each of these pairs of mesenteries are stated by Lacaze 

 Duthiers to be developed synchronously. One might suggest that on the formation of 

 the mesenteries numbered 1, 3, 6, 7, 10, and 12, the ccelenteron is divided into six chambers, 

 one anterior and one posterior, both of which are limited by the directive mesenteries, 

 and four lateral chambers, two on each side of the stomoda3um. At this time evidently 

 the mesenteries which divide the lateral sections into two form a pair. The two lateral 

 chambers become further subdivided by pairs of mesenteries which developmentally are 

 not adjacent mesenteries, but situated on opposite sides of the stomodasum. In this way 

 the lateral chambers become increased from two to five on each side. On such an inter- 

 pretation the arrangement is bilateral, and consists of an anterior and a posterior chamber, 

 together with five pairs of lateral ones. The anterior and posterior pairs of mesenteries 

 consist of two adjoining members, all the others of two opposite members, that is to say, 

 one on each side of the stomodasum. The directives come to be adjacent mesenteries, 

 because no others are added between them. On this hypothesis it is necessary to suppose 

 that in Actiniaria this primitive type of bilateralism has become modified in various ways 

 according to the family. The bearing will be sufficiently evident without further 

 discussion. The precise manner in which it is lost is shown, for instance, for Adamsia in 

 the figures already referred to. It would, at any rate, aid in the explanation of the arrange- 

 ment in Alcyonaria, Edioardsia, Cerianthus, Zoanthus, Madracis, &c. With regard to the 

 Antipathidse it would simplify the interpretation of the mesenteries in the various genera 

 considerably. The three pairs of mesenteries first developed, viz., those termed "primary," 

 are present in all genera. In Cladopathes the development of mesenteries ceases at this 

 point. In Antipathes, Antipathella, Aphanipathes, &c, two other pairs are developed, 

 but never become so important as the three primary pairs. In Leiopathes a further step 

 is reached by the development of a sixth and still more rudimentary pair. Following 



