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



Dendrobraghiid^:. 



I am unable to decide definitely the precise systematic position of the genus 

 Dendrobrachia, owing to the fact that the polyps are badly preserved. It differs most 

 remarkably from all Autipathida? in three or four important points, and must be 

 considered as the type of a new family. Like the Antipathidse it possesses a spinose 

 horny axis, but here the outward resemblance ceases. In transverse section the younger 

 branches show a most exceptional outline. A central canal appears to be absent and the 

 sclerenchyma is arranged in plates radiating from a short central portion (PI. X. 

 figs. 6-8). A faint line may often be noticed running along the centre of each plate, 

 and the horny matter forming the skeleton is deposited in thin lamella? as in the 

 Antipathidaa. The free margin of each plate is dentate or spined. The plates seem to 

 be somewhat irregular in their course, which is mainly longitudinal, and vary in number 

 from four to seven. Lower down, the angle between two plates becomes partly filled 

 up with sclerenchyma, and this process goes on until finally a round spinose axis is 

 produced, not distinguishable externally from that of the Antipathidse. 



The zooids are usually very remote and often arranged in pairs, which are sub- 

 opposite, the base of each extending over half the axis. The zooids do not stand out 

 at right angles to the axis as in Antipathida? generally, but form an acute angle with it, 

 as in many Gorgonidaa. The tentacles, the number of which is uncertain, are not 

 simple fusiform outgrowths of the ccelenteron, but bear a number of relatively elongate 

 branches in all directions. They thus have the dendritic character of the Alcyonaria, 

 and in this respect differ from all Zoantharia, with the exception of the Thallassianthidaa 

 and Sarcophianthidse amongst the Actiniaria. On this account I was at first inclined 

 to regard Dendrobrachia as a genus of Gorgonacea, which approached the Antipathidse 

 in the form of its axis. There appears, however, to be a total absence of spicules or 

 indeed any skeletal structures, with the exception of the sclerobasic axis, which is horny 

 throughout, and deposited in thin lamellae, as in the Antipathidse. Another feature in 

 which Dendrobrachia differs from the Antipathidse is to be found in the fact that in 

 retraction the tentacles may be completely covered by the anterior portion of the 

 body-wall. The fact that nearly every zooid observed was in this condition has tended 

 not a little to prevent a satisfactory examination of their structure. Frequent attempts 

 have been made to ascertain the number and arrangement of the mesenteries by means 

 of horizontal sections, but I have, so far, been unable to arrive at any definite conclusion. 

 In some cases I thought I recognised ten mesenteries, in others eight, and in others 

 again only six. I also failed to recognise any well-marked band of muscular fibres on 

 either side of a mesentery. This genus therefore appears to approach the Gorgonidae 

 more nearly than any other known Antipatharian, and possesses a curious medley of 



