388 ON THE GROWTH OF 



banks ; for which reason pools of water are formed in 

 them, after a continued rain, the only springs and wells 

 they possess. One of the peculiarities of these islands is ? 

 that no dew falls in the evening, that they cause no tem- 

 pests, and do not check the course of the wind. The very 

 low situation of the country sometimes exposes the inha- 

 bitants to great danger, and threatens their lives when the 

 waves roll over their islands, if it happens that the equi- 

 nox and full moon fall on the same day (consequently 

 when the water has reached its greatest height), and a 

 storm agitates the sea at the same time. These islands 

 are said to be also shaken by earthquakes. 



MM. Quoy and Gaimard, in a lately published me- 

 moir, propose, !<$, To examine how corals raise their ha- 

 bitations upon rocks, and what circumstances are favour- 

 able or unfavourable to their growth. %d, To shew 

 that there are no islands of any extent, constantly inha- 

 bited by man, which are entirely formed of corals ; and 

 that far from raising from the depths of the ocean per- 

 pendicular walls, as has been alleged, these animals form 

 only layers or crusts of a few fathoms thickness. 



The following, according to the French naturalists, 

 is the manner in which this addition or superposition 

 of madrepores is effected. In the places where the 

 heat is constantly intense, where the land is indented 

 by bays containing shallow and quiet water, which is 

 not liable to be agitated by great surges, or by the re- 

 gular breezes of the tropics, there also the saxigenous 

 polypi multiply. They construct their habitations on 

 the submarine rocks, envelope these rocks in whole or in 

 part, but do not form them properly speaking. Thus, 



