1907.] AND SUPPOSED SPECIES IN CORALS. 529 



grow downwards : but as a rule I believe those horizontal plates 

 that bear zooids below are but variant forms of growth of other 

 types : for such corals as have no growth-form except that of a 

 flat horizontal plate do not as a rule bear zooids below. 



Some further remarks must be made about plate-like growths ; 

 but here is a convenient place to refer to a rather strange fact 

 that Darwin called attention to, and this is, that vertical plates 

 grow so that they offer their flat surfaces to the currents. The 

 same thing is true of growths that are branching forms, for then 

 the plane of greatest branching is that at right angles to the line 

 of current. At first sight this seems a strange thing, for plates 

 growing in rough water become so much more exposed to damage ; 

 and yet it is doubtless for the exposure of a larger surface of 

 zooids to the food-bringing currents, that the plate spreads in 

 this direction. This phenomenon is especially notable in the 

 Millepores, the broad laminse, or fan-shaped growths of which are 

 always opposed to the line of the waves, even when the colony 

 happens to dwell in very rough water. 



Two general rules may thei-efore be laid down that apply to all 

 the forms of growth that have been described : the first, that all 

 corals having symbiotic algse tend to grow upwards ; the second, 

 that all tend to offer their greatest surfaces to the line of 

 currents. 



We may therefore assume that every coral embryo that settles 

 upon a site of election, and starts the foundation of a colony, has 

 three inherent tendencies : it has its inherited type form of 

 vegetative growth, aiid its inclinations to grow upwards and to 

 oppose its growth to currents. ISTow these tendencies are affected 

 by the nature of the water in which the embryo settles down ; 

 and for nearly every coral there is a modification of vegetative 

 growth dependent on the environment : thus most corals have a 

 deep-water form, a smooth- water form, a rough-water form, and 

 numerous variations depending upon the amount of sedimentation 

 that is taking place in the water of their habitat. A coral that 

 grows in rough water is obviously exposed to injury ; and those 

 people who have stated that the home of election of the corals is 

 the surf-beaten edge of the barrier, have made an error of obsei- 

 vation and of fact. The comparatively lifeless zone of a coral- 

 reef is that part of the barrier that is exposed to the maximum 

 force of the surf, with the rising and falling of the tide. 



Such corals as do grow in the almost perpetual crash of the 

 surf are of a very easily recognised type ; and when contrasted 

 with the forms of the same species that have lived in calmer 

 waters, they appear to have very few points in common. The 

 type of vegetative growth best suited to resist the force of the 

 waves, is of course the i-ounded or flattened massive boulder ; and 

 the Porites masses are the type to which all such growths tend to 

 conform (text-figs. 158, 159, pp. 540, 541). 



This rounded form of growth is brought about in several 

 different ways : in Porites, and other corals, where the equal 



36^- 



