IN THE FORMATION OF CORAL REEFS. 



353 



Previous observations have shown that when a Gorgonian colony is 



removed from its attachment on the reef and allowed to lie on the bottom, 



where it will be moved about by the action of tidal currents or waves (or 



if it is suspended upside down), death follows within a comparatively short 



time. The time necessary for the complete disintegration of the coenenchyma 



of a colony and the consequent liberation of its spicules after it had been 



removed from its attachment on the reef was determined for the 12 species 



which make up the most important element in the Gorgonian fauna of the 



region. The results of these experiments are shown in table 6. 



Table 6.— Time necessary for the complete disintegration of the coenenchyma of 



Gorgonian colonies. 



Under normal conditions on the reef, the greatest number of axial 

 skeletons of dead colonies are found in positions which indicate that the 

 tearing of the colony from its original attachment by wave-action is the 

 cause of the greatest mortality. When destroyed in this manner, the spicules 

 from any colony would be added to the limestone-forming materials on the 

 reef within a few days, at most, from the time the colony was torn from 

 the reef. Another active destructive agent, at least on the reefs about Tor- 

 tugas, is the hydrocoralline Millepora dcicornis. On any reef numerous 

 Gorgonian colonies are found, a part of the axial skeleton of which is incrusted 

 with a growth of Millipora. The work of this destructive agent is particu- 

 larly striking on colonies of Gorgonia flabellum. The axial skeleton of this 

 species still retains its delicate lace-like pattern after it has been overgrown 

 by the white Millepora. In many instances colonies are found in which the 

 basal portion is entirely incrusted by this foreign growth, while the distal 

 portion still bears the normal Gorgonian tissues, apparently as yet unaffected 

 by the destruction of the coenenchyma on the proximal portion. In all such 

 colonies examined the disintegration of the coenenchyma had taken place 

 previous to the overgrowth of the axial skeleton by the Millipora, so that 

 the Gorgonian spicules are set free little by little as the growth of the Milli- 

 pora goes on. Very much the same effect is brought about by the upgrowth 

 of incrusting Bryozoa about a colony. In this case, however, the incrusting 

 organism grows over the surface of the coenenchyma and the death of the 

 latter is brought about on account of the cutting off of the supply of food and 

 oxygen rather than by some toxic substance, as seems to be the cause of the 

 death of the tissues in colonies overgrown by Millepora. The disintegration 



