94 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. [Juiie 14, 



coral, a distinct boundary often more or less encompassed by translu- 

 cent, brown calcareous spar ; intimately a tubular structure, and an 

 occasional isolation from all other organic matter. Without extend- 

 ing the comparison, these diiferences are deemed sufficient to separate 

 the lower greensand fossil from Horner a ; and as similar subordinate 

 tubuli have not been included in the characters of any established 

 genus, it is proposed to make the extinct zoophyte the type of a 

 new one, with the designation Siphodictyum, from aicpojv, tubus, and 

 CLKTvnv, rete, in allusion to the horizontal tubuli terminating in a net- 

 work on the surface. The specific name "^ graciW is retained for 

 the sake of identification with the previous determination. 



SiPHODICTYUM, n. g. 



Gen. char. — Tubular ; the tubes forming branches, and opening only 

 on one side of them ; interspaces and all the reverse surface occu- 

 pied by a network of minor pores ; exterior of branches progress- 

 ively altered by local secretions ; interior occupied by downward 

 prolongations of the visceral tubes, and traversed horizontally by 

 small subordinate tubuli, connected with the surface-reticulation ; 

 additional tubes developed between pre-existing. 



SiPHODICTYUM GRACILE. 



(Pl. Y. figs. 16 to 23.) 



Spec. char. — Branches cylindrical, slender, forked, csespitosely ag- 

 gregated ; surface sometimes traversed vertically by discontinuous, 

 longitudinal ribs ; in aged conditions rugose ; openings of the vis- 

 ceral tubes circular, very distinct, margin sometimes slightly pro- 

 jecting ; order of distribution generally irregular, occasionally trans- 

 verse, where the longitudinal ribs occur, vertically lineal ; minor 

 openings arranged in rows between the ribs, elsewhere retiformly. 



The fine specimen submitted to examination by Dr. Fitton con- 

 sisted of innumerable slender branches, the upper extremities of 

 which presented a caespitose area nearly 2 inches long and 1^ wide. 

 On the side where the general character of the branches was shown, 

 considerable dislocations occurred ; but in one instance was a clear 

 succession of forked branches for li inch. Not the slightest means 

 were afforded for determining the nature of the original base, but 

 it probably resembled that of M. Roemer's ChrysaoraX pulchella 

 {op. cit. tab. 5. fig. 29 c), or a solid disc, from which sprang single, 

 closely aggregated stems. The branches were simply and in general 

 uniformly forked (fig. 16) ; and the plane of separation was almost 

 invariably the same in successive bifurcations. The intervals between 

 the points of subdivision were very irregular, varying from f ths of a 

 line to 3 lines. The differences in the dimensions of a branch due to 

 thickening were small, the breadth of the younger conditions being 

 about f ths of a line, and that of the older half a line. Most frequently 

 no increase preceded a bifurcation ; and sometimes no difference was 

 perceptible between the undivided branch and each of the shoots. 



Notwithstanding the great extent of the specimen, a very young 



