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



There are from about fifteen to twenty-two dactylopores in each cyclo-system, in form 

 and arrangement almost identical with those of Astylus subviridis. 



The margin of one side of each cyclo-system is raised up into a stout projection, 

 which is inclined slightly over the mouth of the system for a short distance. After 

 running this inclined course the projection spreads out into a thin broad lamina, with 

 a rounded border, which extends horizontally over the mouth of the cyclo-system, and 

 hangs as a lid or cover over its entire extent (PI. II. fig. 7, and PI. XII. fig. 7, a, b). The 

 inclined portion of the projection is stout and thickened, and is strengthened by being- 

 continuous at its base with the adjacent ccenenchym of the branch. It is thickened to 

 the greatest extent in the direction towards the centre of the cyclo-system, and so much 

 so that its substance projects within the cavity of the upper chamber of the gastropore as 

 a prominent ridge. This ridge, becoming gradually less marked as it descends, is con- 

 tinued downwards to the margin of the aperture leading from the upper to the lower 

 chamber of the gastropore, and appears as a prominent thickening of the wall of the 

 upper chamber in this region (PI. II. fig. 7, A). The dactylopores are aborted and 

 absent in the region of the cyclo-system overgrown by the base of the lid. The lateral 

 margins of this base are often grooved by dactylopores on either side, which have the 

 appearance of having been pushed aside, as it were, by the growth of the projection. 

 The thin horizontal lamina constituting the lid of the cyclo-system is often not quite 

 smooth in surface, but somewhat undulate or crumpled, as it were. The lids are all 

 directed with great regularity towards the tips of the branches on which the cyclo- 

 systems to which they belong rest ; the supports of the lids arising from the sides of the 

 cyclo-systems nearest the origins of the branches. 



The gastropores are divided into two chambers as in Astylus. In the present form 

 the lower chamber is relatively smaller than in Astylus subviridis. Its communications 

 with the dactylozooids are closely similar to those in Astylus (PI. II. fig. 7). The 

 opening between the two chambers in Cryjitohelia is circular, not horse-shoe shaped as 

 in Astylus. 



Ampullae occur only in connection with the cyclo-systems in Cryptohelia pudica. 

 In the female specimens examined by me, only one ampulla is developed in connection 

 with each system. It may lie on either side of the system, but not on the back 

 of the flabellum. The ampullae are rounded cavities of irregular form, which, when 

 mature, are so large as to be as wide as the side of a cyclo-system, and occupy it entirely 

 (PI. XL). 



Numerous specimens of Cryptohdia pudica, the soft parts of which are not preserved, 

 have several ampullae developed in connection with each cyclo-system. These are 

 concluded to be male examples. The ampullae are not nearly so large as in female 

 specimens, and do not give evidence of their presence by forming swellings on the 

 surface of the corallum. 



