32 Coelenterata. 



The septa of E. unite at the centre to form an incomplete or irregular pseudo- 

 columella. In the lower Devonian E. strictum the pseudocolumella is further 

 developed, moreover the tertiary septa, which in all the earlier genera are free, 

 in this and most later genera are attached by their inner margins to the next- 

 adjacent dorsal primary or secondary septum. In the middle Devonian begins 

 a rapid divergence of various specialised lines of development. In the direct 

 line from Str. is Stereolasma rectum, with a complete pseudocolumella, rein- 

 forced by a deposit of stereoplasm between the septa and near the pseudo- 

 columella. Other lines led to Heterophrentis, to Heliophyllum halli and con- 

 fluens, to Microcyclus and Hadrophyllum. The acme of development of the 

 Rugose corals, and of the Str. group, was reached and passed in the De- 

 vonian, with the beginning of the Carboniferous only a few terminal members 

 of the various series are left, and at the close of the Carboniferous the whole 

 group was extinct. Lophophyllum was apparently derived from Stereol., and 

 Hapsiphyllum from Heterophrentis. 



Grabau states that palaeozoic corals show, in their septal development, 

 a fundamental tetrameral plan. This is persistent in the earliest known genera 

 but becomes masked in later species by the secondary assumption of radiality. 

 The development of the mesenteries of modern Hexacoralla shows a similar 

 order of appearance, the 1st and 2d pairs are ventrads (with muscle ridge 

 turned ventrally), the 3d (ventral directive) is a pair of dorsads and the 4th 

 (dorsal directive) a pair of ventrads. The 5th and 6th pairs are dorsads form- 

 ing, with the 1st and 2d pairs, 4 false pairs of braces*. After this the 

 mesenteries appear in compound pairs, a pair of dorsads and a pair of 

 ventrads simultaneously; thus in the corresponding intermesenterial spaces a 

 brace of new mesenteries appears, the order being comparable, even in detail, 

 to that of the septa in the Palaeozoic Tetracoralla. 



Carruthers( 1 ) traces the evolution, from Zaphrentis delanouei, of 4 new species 

 in ascending horizons of the Carboniferous Limestone series, and points out 

 the changes which occur in the shape of the cardinal fossa and the length of 

 the major septa. 



Taylor gives an account of the geographical distribution, morphology, form 

 variation, growth change, colonies, budding, the structure of the wall, septa, 

 tabulae and synapticula, the chemical composition of the fossils and microscopic 

 characters of the skeleton of Archaeocyathinae. The most extensive series are 

 known from South Australia and Sardinia. The Arch, have been found at many 

 points along the coasts of the Cambrian continents from 70 N to 30 S. The 

 organism was attached to the substratum by a mass of calcareous lamellae or 

 by cylindrical root-bundles; the author opposes the >free calyx theory of 

 Bornemann. The organism at first builds a hollow cone, but this may develop 

 into a cylinder, turbinate cup, or broad, circular or oval funnel, and examples 

 with hour-glass section are common. Several species build an open saucer- 

 shaped skeleton, and flabellate expansions or large folded bowls also occur. 

 The structure of the cup walls is similar throughout, there is a distance of 2 to 

 3 mm. between the outer and inner walls. Sections gradually approaching the 

 tip show that the number of septa decreases to six. Several specimens exa- 

 mined indicate that the Arch, build a basal plate upon which the outer wall, 

 6 septa and inner wall arise simultaneously. The class is divided into 5 fa- 

 milies Archaeocyathidse ArchcBOcyathus 11 n., Ethmophyllum 1 n., Ar- 

 ch&ofwngia n. 1, Pycnoidocyathus n. 3, Somphocyathus n. 1; Coscinocyathidae 



Coscinocyathus 4 n., Ooscmoptycha n. 2, Archaosycon n. 1; Dictyocyathidae 



Dictyocyathus 2 n., Dokidocyathus n. 1 ; Spirocyathidae - - Spirocyathus 2 n., 



