1905.] CAPE VERDE BIARINE FAUNA. 185 



shallow beds of limestone which clothe the lowei- slopes of the 

 sides of these two valleys, and this is deposited as the cement of 

 the sandstone when the fresh water mixes with the salt. 



But for the interference of tides and waves the level of this 

 rock-bed would be that of the valley, i. e. the top of the sand- 

 beach, where indeed a very soft rock does occur. But the sea 

 has cut down this mass to the level determined by the height of 

 the tides &c. in the usual way. As in the case of the reefs of 

 Zanzibar, we are here shown how purely physical causes, aided 

 by protective organic growths, can jaroduce reefs closely i-esem- 

 bling those the mass of which is due to growth of organisms 

 in siiiij, which leads to the conclusion that the forms characteristic 

 of true coral-reefs are very lai-gely due to the physical action 

 of the sea as well as to the laws of growth of the organisms 

 themselves. 



That the formation of this sandstone-rock is still proceeding 

 seems most probable, and it would be interesting to know defi- 

 nitely the conditions of the landward side of the strata underneath 

 the alluvium of the valley. The solvent and eroding actions of 

 the sea are very nearly balanced by the growth of calcareous 

 organisms, but the presence of outlying rocks and submerged 

 masses indicate that the sea is slowly encroaching. It is probable 

 that cementing of fresh material is being carried out on the 

 landward side of the strata at the point where fresh and salt 

 water meet — a point which is moved backwards just as fast as the 

 sea encroaches. 



It is to be noted that the occurrence of beach -sandstone is 

 not confined to valley- mouths. Smaller deposits. -ogeuPelsewhere, 

 and a sample I collected at the'.' north "p)int of St7 Antonio has 

 the appearance on the surface and the extreme hardness charac- 

 teristic of the East- African " Coral-rag." 



6. Summary and Conclusions. 



1. Faunistic. — Although the Indo-Pacific Oceans are one faun- 

 istic area, there is no fauna common to the tropical seas of the 

 world, so fai- as the evidence of the Cape Verde Islands goes, 

 though certain species are common to the Ti'opics of both Atlantic 

 and Indian Oceans. If the Tropical Atlantic is a distinct area 

 characterised by a special fauna, its northern limit is carried far 

 to the south by cold currents (and probably northwards in the 

 south), as in spite of the position of these Islands, between 17° 

 and 15"^ N.*, their fauna has a considerable constituent derived 

 from the subti-opical zone. But it may later appear that this 

 portion of the favma is unduly conspicuous through its sj)ecies 

 being already well-known, while the constituent derived from the 

 Tropical Region, if any, will not be known until the systematic 

 examination of the collections is completed. 



The scanty representation of some groups (e. g. Corals and 



* Compare the fauna of Suez Bay, which is tropical, though situated in latitude 

 30° N. 



