30 Coelenterata. 



the general surface of the fractured end showed 70 newly formed coralla of 

 old and new zooids. During the same interval of time a similar branch on 

 the same colony, which had received no injury, had advanced 1,5 cm. and had 

 added about 120 new lateral zooids. The rate of repair is thus a rapid one. 

 If the apical zooid be destroyed, its functions are taken over by the more 

 vigorous lateral zooids so that there is a tendency to branch formation below 

 the site of the injury. A coral may repair an injury by a new growth of a 

 different type to that of the colony, and in such cases the repair growth is of 

 a form better suited to the environment than is the form of the parent colony ; 

 even the type of the corallum may be changed, e. g., a Montipore, instead of 

 showing a smooth surface over which the fairly wide mouths of the corallites 

 are dotted, shows, in its repair growth, a surface with numerous papillse at the 

 base of which open the minute corallites. Silt, sand, or suspended matter may 

 cause the death of corals (1) by falling on them from above or (2) by over- 

 taking them from below. Rough-water forms of Madrepora and Poc. were 

 transported, without injury or delay, from their habitat of rough barrier water 

 to a sheltered sandy pool of the barrier flats, most of them died within a month 

 and all were dead in 50 days, the death being due to silting. It is just 

 on account of silt that many acres of the lagoon are devoid of coral growth 

 and probably on account of the presence of silt that wave action is so necessary 

 to coral life and that the unstirred depths below about 20 fathoms are com- 

 paratively bare of coral. There is no species of coral in this atoll which is 

 not able to withstand an exposure of many hours to the mid-day sun, with 

 from 6 to 12 inches of its apical growth above water; all the barrier species 

 normally endure this at mid-day spring tides. Exposure to sun and air, between 

 tide limits, plays little part in causing the death of corals. The author believes 

 that Millcpora alcicornis, complanata and verruoosa are merely variations of 

 vegetative growth as there are gradations connecting them. Probably the 

 numerous forms of Pocillopora are also variants of a single species. See 

 also Bedot, Gravier( 8 ), Chadwick, Hickson( 2 ), Vaughan(V). 



According to Gravier( 4 ) West Africa does not appear to be as devoid of 

 coral formations as Darwin believed. Poritids and Astrseids occur at St. Thomas 

 (Gulf of Guinea) but do not form reefs (in the usual acceptation of the term). 

 The coral growths are not exposed at any state of the tide. See also 

 Gardiner and Vaughan( 3 ). 



Brown, on investigating the Devonian coral Streptdasma rectum, finds at the 

 tip of the corallum 4 primary septa. The alar septa are inclined to the 

 cardinal septum so that the counter quadrant spaces are larger than the cardinal 

 ones. Then 2 secondary septa appear in each counter quadrant, these are 

 distinctly smaller than the primary septa and are not radially placed but joined 

 by their inner border to the dorsal side of the alar septa; in some specimens 

 these septa later become equal in size to the primary septa and radially arranged. 

 Duerden [s. Bericht f. 1906 Coel. p 31) is in error in claiming primary hexa- 

 merism for all Rugose corals, for in this one there is a stage with 4 septa 

 and as the third pair of septa appears later and develops in the same way as 

 the other metasepta, being inclined to the alar septa, they are to be regarded 

 as metasepta. In S. rectum, the counter quadrants are accelerated as compared 

 with the cardinal ones, for a tertiary septum appears on each side of the 

 counter septum (dorsal directive), that is in each counter quadrant, before even 

 a secondary septum has been formed in the cardinal quadrants, and 3 secondary 

 septa have appeared in each counter quadrant before the first secondary one 

 appears in the cardinal quadrants. The possibility of proving primary hexa- 



