TJIE BEANCHING-SVSTEMS OF THE ALCTONACEA. 533 



Further, I have not beeu able to discover that the lateral members of these three 

 species ever branch on the principle of a monochasial cyme, as is the case in L. nitkla. 

 Sections through branclilets of L. rhahdota and L. perlstyla show that all the lateral 

 members communicate, either directly or indirectly, through the intermediary of lateral 

 members of the first order, with the chief member forming the axis of the branchlet. 

 Thus in these species the mode of branching is that of a compound raceme, without 

 additional complexity. It appears that the more comjjact arrangement of the anthocodise 

 on the branchlets of these three species is due to their simpler mode of branching as 

 compared Avith L. nitida. 



I have entered into these questions at some length because the mode of aggregation 

 and the system of branching in the Alcyonaria have been insufficiently studied and are 

 imperfectly understood. 



Botanists have provided us with a terminology apj)licable to every conceiA'able mode 

 of branching, and it is possible, by making use of their system, to throw light on many 

 obscure points in the .Vlcyonaria. 



Anatoxin i)f the Zoo'nh. 



Fig. 18 (PI. 42), is a drawing of a group of terminal anthocodiicfrom a branchlet of Z. nit Ida 

 which has been decalcified, stained in alum carmine, and made transparent in oil of cloves. 



The tentacles are sliort and broad, provided with a single marginal row of pinnu.les, 

 their bases confluent so as to encompass a capacious hollow, the oral disc, into which the 

 tentacles can be infolded. The mouth-opening is circular, rather large in L. nitida, 

 small in L. p^i-isti/la ; the stomodiBum long and pyriform, its smaller end directed 

 downward. The ciip-shaped oral disc and the upper wide region of the stomoda3um are 

 lined by a tall columnar epithelium containing a few gland-cells, but no nematocysts, so 

 far as I could see. I could not detect any trace of cilia on the epithelium of this region. 

 Tlie uppermost swollen part of the stomodaeum is oval or subcircular in section, and is 

 not so much compressed from side to side as is usual ia the Alcyonaria. There is no 

 sulcus in its ujipor part, but towards its narrower end the cells lining its outer or abaxial 

 wall change their character, become attenuated and elongated, and bear each a single 

 very long cilium. This strip of modified epithelium forms a shallow gutter which 

 is undoubtedly a sulcus, but further down, in the nai-rowest part of the stomodseum, 

 a definite sulcus can no longer be spoken of. In this r(>gion the stomodfeal tube 

 is compressed laterally, and is lined throughout by a modified ciliated epithelium 

 resembling that of the sulcus. In some cases the sulcar groove disappears entirely, and 

 the cavity of the stomodaeum is a simple oval as represented in fig. 21 ; but more 

 frequently the cells on the long sides of tlie oval project further into the cavity than 

 those at the two narrower ends, so that the section is keyhole-shaped with a groove at 

 either end. But one cannot speak of a sulcus and sulculus in this case, for the epithelium 

 luiing the tube is of the same character throughout. Such an arrangement has not, so 

 far as I am aware, been described in any other Anthozoiin *, and it may be regarded as a 



* Ashworth, Proc. Eoyal Society, Ixiii. (1^9^i) p. 443, describes the stomodieum of Xenia as having a well-inarkcd 

 siphonoglyphc (sulcus) in which the ceils ot the lower tliird only bear long fiagella. 



