1038 MISS R. HARRISON ANB PROF. S. J. HICKSON ON 



Wcas broken by tlie formation of the Isthmus of Suez. But the 

 last connexion but one was that by way of a strait, of which 

 the remaining part is now the Persian Gulf ; and this gulf with 

 its narrow outlet into the Indian Ocean, high temperature and 

 great rivers, might be expected to i-etain some of the fauna which 

 had been subject to veiy similar conditions in the South-east 

 corner of the Mediterranean Sea. Melvill and Standen call 

 attention to the considerable generic analogy between the Mollusca 

 of the two regions, and point out that the species of the 

 Persian Gulf show close affinities with South-European forms. 



It would be quite pi'emature to draw any far-reaching 

 generalisations as to the distribution of Madreporaria from the 

 few specimens that are here recorded, but attention may be 

 called to one or two points of genei'al interest. 



The occurrence of Pyrophiillia in deep water in the Persian 

 Gulf is interesting from the point of view of geographical 

 distribution, Alcock [3] in his comments on the deep-sea Madre- 

 poraria of the Indian Ocean, calls attention to the " many 

 intimate affinities of the fauna of moderate depths of the Indian 

 seas with the North Atlantic faui:ia," and considers them "to be 

 sufficient to suggest a direct sea-connexion, in the past, between 

 the Atlantic and Indian Oceans, and the case of Caryojihyllia 

 communis and Flabellum ladniatum w^ould indicate that the 

 connexion was by way of the Mediterranean." 



The case of Pyropliyllia and Giiynia appears to me to give 

 even stronger evidence of the truth of this hypotliesis, than that of 

 the two species quoted. The' genera Garyophyllia and Flahellum 

 are both very widely distributed recent corals, and it is possible 

 that in comparatively recent times these two species may have 

 had an almost cosmopolitan distribution. Guynia and Pyrophyllia 

 are, so far as is known at 'present, very restricted in their distri- 

 bution and are totally unlike any other recent coral — with the 

 possible exception of the West Indian genus HaplophylUa. 



Although so much alike in important characters, they are 

 sufficiently distinct for us to believe that they were separated 

 from one another at a verj^ remote period. It is rather more 

 difficult to believe that the Indian Ocean and Mediterranean 

 specimens of Caryop]t,yUia comm,unis aiid Flahelhmn laciniaium 

 can have undergone no differential change since the time when 

 the Mediterranean Sea and the Indian Ocean were in commu- 

 nication. A second point of interest is, that the only other coral 

 with which Pyrophyllia shows affinities, namely Conosmilia, 

 should be found in the Tertiary deposits of Australia. Standing 

 by itself, this is only one of those facts of geographical distribution 

 which it is important to note but impossible to explain in a 

 satisfactory way. But its importance as a fact is emphasized 

 when it is placed side by side with the facts of the distribution of 

 Trematotrochus. 



As Miss Harrison points out, the specimens of Trematotrochus 



