100 On the Structure of the Siylasterida. [June 15. 



inclination of the contained tentacular zooid when at rest, or when feed- 

 ing the deeply seated alimentary zooid. The supposed included septa of 

 higher order are the styles of the tentacular zooids. In some forms of 

 the family these styles are brush-like in shape, just Kke the central styles 

 of the alimentary zooids ; they have this form in Allo]pora mineacea *, 

 and less markedly in Stylaster comj^lanatus, Pourt. t In some Stylas- 

 teridae, as e. g. in Stylaster amphihelioides, S. Kent t, there is no appear- 

 ance at all of pseudo-septa. The pores of the tentacular zooids are simple 

 circular-mouthed pits, arranged in a circle around the large pore of the 

 alimentary zooid. In Allopora suhviolacea, S. Kent §, the pores of the 

 tentacular zooids are, in some zooid groups in the same specimen, mere 

 pores ; in others slits communicating with the cavity of the pore of the 

 ahmentary zooid. The irregularly scattered condition of the zooids 

 existing in Polyjpora is to be regarded as the primitiTe one in genesis, 

 from which that existing in Stylaster ampJiihelioides and that in Allopora 

 suhviolacea represent transitional stages towards the high specialization 

 of the zooid groups found in Cryptohelia and other species at present 

 termed Stylaster. 



It has hitherto been a matter of regret that the Hydroidea were of such 

 a structure as to be unsuitable |j for preservation in the fossil state, and 

 that thus we were almost, excepting as far as Grraptolites are concerned, 

 "without direct evidence as to the forms which may have been presented 

 by theii' remote ancestry. We have now two families excellently adapted 

 for preservation as fossils, viz. the IMilleporidae and the Stylasterid^. At 

 present no members of these families appear to have been observed in 

 rocks older than the tertiary deposits. A single species only, Disticliopora 

 antiqua, is known to occur in tertiary beds in France, at Chaumont and 

 Yalmondois ^; but now that special attention will be directed to these 

 corals, and their structure is better understood, no doubt allied fossil forms 

 will be detected. It seems just possible that amongst Palteozoic corals such 

 forms as Cyatlionaxia may have been tenanted by a group of hydroid 

 zooids with a large alimentary zooid situate upon the projecting style. 

 Cystlphyllum vesiculosum has a crowd of small slit-like pits covering the 

 inner surface of its calicle, which have all the appearance of ha\ing been 

 tenanted by hydroid tentacular zooids. I cannot, however, now refer to 

 specimens ; indeed I have never seen any. Ampulla seem to be absent 

 in these corals ; but in shallow-water forms, as in Millepora, they proba- 

 bly would be so. It is quite possible that the IMillepores produce 

 Medusae, 



Although the Milleporidas take a very large part in the formation of 

 coral reefs, the St3'lasteridae have very httle share in the building up of 

 these structures, being for the most part confined to the deep sea. A few 



* Pourtales, I. c. pi. iii. fig. 15. t Pourtales, I. c. pi. ii. fig. 17. 



+ Saville Xent, I c. pi. xxiv. fig. 1 c. § Ibid. pi. xsv. fig. 2 a. 



1! Allman, I. c. vol. ii. p. 231. 



^ MM. Milne-Edwards & Haime. /. c. t. iii.. Appendice, p. 451. 



