1866.] GIJPPY WEST-INDIAN TERTIAKIES. 573 



them in the Tamana series all belong to the same horizon as the 

 Manzanilla beds. There does not, however, appear to be any reason 

 on stratigraphical grounds why these latter, with the San Fernando 

 beds, should not be older than the other Tertiaries of Trinidad. 



Pig. 3. — Section from Nariva Swamp to Mount Calabash^ 

 3| miles from the coast (6 miles). 



S.S.W. Mount Eiver Mount N.N.E. 



Nariva Swamp. Lebranche. Lebranche. Eiver, Calabash. 



^ c b a 



a. Clays and shales (Caroni series). 



b. Clays, calcareous sandstones, &c. (Manzanilla beds). 



c. Sandstone and indurated clays (Neocomian). 



Their organic contents seem to consist of a different assemblage of 

 species to that found in the other Miocene deposits, containing at 

 the most, out of more than thirty species of MoUusca collected by 

 me, but four found elsewhere. Consequently I place these forma- 

 tions, provisionally only, in the Lower Miocene until further evi- 

 dence may be forthcoming as to their true age. 



The fossils from the Manzanilla beds which are now described, 

 consist of the following species : — 



Cercomya ledseformis, spec. nov. 

 Erycina tensa, spec. nov. 

 Area Trinitaria, spec. nov. 



filicata, spec. nov. 



Cardium castum, spec. nov. 



Ancillaria laniellata, spec. nov. 

 Venus Walli, spec. nov. 

 Dosinia cyclica, spec. nov. 

 Mactrinula macescens, spec. nov. 

 Corbula vieta, spec. nov. 



Of the above, the Ancillaria somewhat resembles A. glandina of 

 the Paris and London basins. Dosinia cyclica is a remarkable form, 

 connecting the genera Dosinia and Cyclina. Mactrinula macescens is 

 distantly related to M. prcetenuis (Conrad) of the American Eocene, 

 and more closely to M. plicataria, a hving species of the Indian seas. 

 Corbula vieta occurs in the Miocene of Jamaica and the Pliocene of 

 Trinidad. It is alhed to O. gibba and C. pisum of the Hampshire, 

 Isle of Wight, and Belgian Tertiaries. Erycina tensa strongly re- 

 sembles a Paris-basin species. With regard to Cercomya ledoiformis, 

 if I have rightly determined the generic relations of that shell, it 

 will be the first recorded appearance of the genus in beds of so late a 

 date. Among the undetermined shells are a Strombus and a Pleuro- 

 toma, which are possibly identical with species from San Domingo 

 and Jamaica ; but as the specimens are imperfect I do not venture 

 to name them. 



b. Anguilla. — The Miocene strata in this island consist of a white 

 marly limestone containing a large number of MoUusca and Echino- 

 dermata, the former mostly in the state of casts and specifically in- 

 determinable. The Echinodermata are better-preserved. I have no 

 data as to the position of the beds ; but it is worthy of remark, in 

 respect to the influence of the conditions of mineralization upon the 

 remains preserved in the rocks, that in Anguilla, where the deposits 



