9G MADREPORARIA. 



84. Porites West Indies o: 19. (P. Americana incertn: aedis nonadecima.) 

 (ri.V. fig. 8; PI. XVI. fig. 1.) 



[ British Museum.] 



Drsrripfion. — The coralhnn ri.ses from a IhiiKiblique stem (? overturned) from 1 to 1-2 cm. 

 tliick into a rapidly expanding cluster up wavy, forking, and gradually thickening stems ; 

 it is some 12 cm. high and 15 cm. in diameter of its nearly level top. The great number of 

 undivided stems are due to rapid forking, without trace of regular dichotomy. They are very 

 crowded but free, that is, do not fuse. In thickness they vary greatly, and wave about, some- 

 times digitiform, sometimes flabellate or irregularly swollen. The living layer is confined to 

 the tips, is only 2 -5 cm. at the deepest, and each separate tip is either a small single knob or 

 swelling, flattening, and dividing quite irregularly, putting out divisions almost wherever there 

 is room for tliem. An epithecal film creeps up over dying edges. 



The calicles are small, average about 1 mm. in diameter, and shallow, though with 

 a skeleton sufficiently open as to appear deep. The wall over all the tips is delicately 

 membranous, and raised cup- or funnel-shaped above the surface ; lower down it is a thin w^^ll- 

 tliread, here and there becoming reticular, with fiaky filamentous elements and smooth round 

 pores or meshes. The septa are thin and delicate where they actually leave the walls, but 

 soon swell into gi-anules. They are very perforated, and lower down project as echinulate rods 

 joining the pali and the columellar tangle ; seen from above, the typical septal formula can be 

 made out ; seen close, the skeleton is too broken up into an apparent confusion of frosted or 

 echinulate granules to show any symmetry. The pali rise up from the tangle as very small, 

 irregularly star-like granules, varying in numlier according to the development of the directives 

 from five to seven. There is usually a smaller central tubercle rising with the pali from an 

 open, somewhat scanty, central tangle. The tubercle rises at times from systems of spokes of 

 t his tangle. The interseptal loculi, in spite of the echinulate sides of the septa, are large and open. 



In section, the axial strand is of stout streaming network, showing the outlines of calicles 

 very distinctly; it is not veiy extensive, and changes gradually into a stout, radially streaming 

 reticulum, in which trabecule are fairly conspicuous. 



The colour of the unbleached coral is a rather deep warm buff. 



This is another of the West Indian Poiifes in which the uprisings tend to enlarge 

 as they rise, causing great irregularity in tlu^ ordinary dichotomous branching, see Table III. 

 ]). 136. The great width of the cluster rising from so small an attachment is interesting. It 

 must have very soon become top-heavy. All the knobs into which its colony is broken up 

 reach about the same levijl. The method of growth should be compared with that last 

 described. It is veiy different from either the clavaria or furcata of Lamarck, while again, 

 they differ from one anotlun- by characters wliich most systematists, working on the ordinary 

 lines, would claim to be specific. 



a Large stock. I'.tOG. 1. 1. 4. 



b Broken fmiii n, and with two fra"ments, one of which is lileacbed. 1906. 1. 1. 5. 



