ON THE GEOLOGY OF HAKBADOS. 229 



slrengtliens the supposition tliat the liarl);i(lian rock is of ^fiooene 

 ago. Plot'. J)uiu-ai» nuMitions two other corals from J5arbados, but 

 they were not speciticnlly recognizable, and it was uncertain whether 

 they had been obtained from the raised reefs. The occurrence of 

 Astraa (now called JJeliciatnca) harhadcnsis was then the sole link 

 between the reefs of Barbados and Antigua, but the connexion is 

 now strengthened by Mr. Gregory's discovery of Heliustrcpa erassi- 

 Jamellata in the higher reefs of Barbados. 



In respect to Barbuda, the only coral yet obtained from that 

 island is the species described by Prof. Duncan under the name of 

 Cyphastnca costata. This has now been found in the lower reefs of 

 Barbados, and will doubtless be discovered in Antigua and Guada- 

 lupe when those islands are more thoroughl}^ searched. Prof. Duncan 

 also records it from San Domingo and Jamnica, so that it constitutes 

 another link in the chain of evidence that the collections examined 

 by Prof. Duncan included a mixture of Pleistocene, Pliocene, and 

 possibly some Miocene species. A species of Cyphastrcea, C oblafa, 

 still lives in the Caribbean Sea, and others occur in the Pacific ; but 

 Prof. Duncan considers that the fossil form differs from all of them. 



Barbuda being a small low island, and the shells in the coral- 

 rock being all of recent West-Indian species (see p. 231), there is 

 really no ground for regarding as Miocene the bed from which the 

 Cyi>hastr(ea came. The fossil indeed is expressly stated to have 

 been obtained " from the hard superficial limestone." 



Finally, with regard to the alleged resemblance of the fossil corals 

 with recent Pacific species, such affinities (admitting that they exist) 

 would be completely accounted for if there had been free communi- 

 cation between the Pacific Ocean and the Caribbean Sea up to a 

 late Pleistocene date, and, as we shall show, there is a high proba- 

 bility of such communication having existed. 



Prof. Duncan truly observes that the gradual upheaval of the 

 former coral-reefs and banks must have caused a vast alteration in 

 the physical geography of the West-Indian seas, and that " the area 

 of elevation was a vast region ; " yet he still infers that this up- 

 heaval terminated the Miocene age in the Caribbean region, and it 

 does not seem to have occurred to him that the Pacific affinities and 

 the general Miocene aspect of the coral faunas which are preserved 

 in the later Caribbean Teitiaries might be due to the peculiarity of 

 tlie geographical conditions which preceded this immense physical 

 change. 



§ 6. Comparison of the IIaised Keefs of Barbados with simiiar 

 Formations in other West-Indian Islands. 



It is now well known that raised reefs and coral-limestones, 

 similar to those of Barbados, occur in many of the Antilles, but 

 oiir information regarding the structure and fossil contents of these 

 coral-rocks is still meagre. W^e have gathered such information as 

 we could regarding them, the very deficiencies of which will show 



