440 PROF. P. M. DUNCAN ON DEEP-SEA [May 16, 



Cylicia tenella, Dana, which is placed amongst the Astrangiacese ; 

 but the following structural peculiarities render this a matter of 

 doubt. 



1. Some corallites attain the height of T 4 y inch nearly equalling 

 10 millimetres. 



2. All the corallites do not arise from the stoloniferous base, but 

 some spring from others between the calice and the base ; but it does 

 not appear that these are buds. They are probably the result of 

 polypes that became fixed on to the epitheca of the original one in 

 their mobile stage. 



3. The epitheca is shown perfectly in small corallites ; but the 

 larger and even some of the smaller have distinct but small costse, 

 which are marked with granules, and which join the angular and 

 very slightly exsert ends of the septa. 



4. The septal arrangement is irregular. There are six primary 

 septa, and a number of others, the majority resembling the prima- 

 ries. Usually there are 20 large septa reaching into the calice ; 

 and between each pair there is often, but not invariably, a rudimen- 

 tary septum. In some calices the primaries appear to be five in 

 number ; and in a bud there is one primary and a small septum on 

 either side of it. All are granular and incised more or less. 



5. The endotheca is absent, except within the calice of two spe- 

 cimens, where it is extending between the septa near the wall, as if 

 commencing to occlude the floor of the calice. 



6. The columella is formed of processes from the septal ends, is 

 small, and has three or more beautiful granular ornaments on the 

 points where the reticulations meet. 



The incised nature of the larger septa along their slanting internal 

 margin is evident. Sometimes the septa do not even reach to the 

 calicular margin ; and some appear as spines from the inner part of 

 the calicular wall. The third cycle is rarely complete if the rudi- 

 mentary septa, which project between the larger J-^ inch or a little 

 more, are wanting, as in the diagnosis of Milne-Edwards and Jules 

 Haime, ' Hist. Nat. des Corall.' vol. ii. p. 608 ; and the third cycle 

 do not always bend towards the secondary septa. If the rudimentary 

 septa are counted, the fourth cycle is rarely complete. 



The species Cylicia tenella and C. verreau.vi, Ed. & H., differ 

 in respect of the perfection of the third and imperfection of the 

 fourth cycle, this last never being perfect in either. There is no 

 specific distinction between them ; and therefore the last-named 

 species had better disappear. The form under consideration I deem 

 a variety of Cylicia tenella, var. natalensis (Plate XL. fig. 3). 



Madreporaria perforata. 



Family Exjpsammin^e. 



Balanophyllia helen^, sp. ii. (Plate XLI. figs. 5, 6.) 



The corallite has a small base, a narrow bent cylindrical stem, 

 and a suddenly large calicular opening. 



The costse are not exsert at the calice, are distinct to the base, 



