MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 489 



-All stages in the resorption of the lower parts of the mesenteries can be observed. Where 

 the action is in progress the peripheral edge is free and tapering, although the mesentery was 

 originally attached by this to the wall (fig. 158). The mesoglcea is seen to break up into distinct 

 pieces, and the endodermal epithelium is in different stages of disorganization; terminal 

 Fragments appear as if about to break off, and occasionally free particles are met with. The 

 mesenterial debris thus set tree is evidently injested by the endodermal epithelium lining the 

 chambers, for the layer is here of exceptional thickness and the large cells are crowded with 

 granules and irregular fragments, which closely recall those given off from the disintegrating 

 mesentery. 



Fig. 153, taken from a retraeted polyp, reveals that the mesenteries do not extend as far as 

 the most peripheral chambers, although the region represented is no lower than the stomodseuni. 

 In some cases a fragment of the mesentery may persist in the second chamber, but its imperfect 

 character indicates that it is about to disappear; even where the section does not actually 

 encounter a synapticular interruption the peripheral tissue is atrophied. 



In the tangential section, fig. 156, the mesenteries all extend vertically beyond the first 

 transverse rows of synapticula, but in the chambers below they begin to exhibit the various 

 stages in absorption. 



That the mesenteries are actually pierced by the synapticular formations is manifest from 

 the preparations. "When serial sections are passed in review, it is seen that the mesentery 

 wholly surrounds the upper and more central perforations left by the removal of the synapticula, 

 and frequently the mesenterial mesoglcea becomes swollen, and presents striated areas, such as 

 are formed by the desmocytes where a mesentery is inserted on the calicular wall (fig. 157). 



Miss Ogilvie has attributed an altogether different origin to the synapticula, in her account 

 of these structures in Fungia and Siderastrsea. Commenting (p. 170) upon Bourne's description 

 of the synapticula in Fungia, she states: "The important point is that they neither 'interruptf 

 nor 'jiirrcc' the mesenteries." Further, it is assumed all along that the body wall is specially 

 invaginated from below to produce them". Had an examination of the actual poly pal tissues 

 been made it is impossible to see how any support could have been adduced for such statements, 

 any more than would be forthcoming for the production of simple tubercles on the septa. 



Professor Bourne, in his paper. "The Anatomy of the Madreporarian coral Fungia" (1887), 

 also describes somewhat similar mesenterial relationships in the genus Fungia, only here the 

 synapticula are in single vertical or oblique bars, not in vertical rows, as in Siderastrsea. In the 

 upper regions of the interseptal chambers there are no synapticula, and the mesenteries are free 

 to radiate across the whole space between the stomochvum and the periphery of the disk, but in 

 the lower portions of the loculi the continuity of the mesenteries becomes interrupted by 

 the synapticula. Owing to the much larger number of vertical bars across the broad' septa 

 of Fungia, the intersynapticular cavities in sections greatly outnumber those of Sidei'astrasa, 

 and the mesenteries do not extend wholly across any segment, being represented by a small 

 projection at each extremity of the chamber (Bourne's figs. 13, 15). Bourne's explanation 

 (1887, p. 19) of the significance of the synapticula. that "physiologically they seem to serve 

 as stays or buttresses, giving solidity and coherence to the corallum," is probably the most 

 correct of any vet offered. From the disappearance of the mesenteries below, almost paripassu 

 with the development of the synapticula, the circulation of the digestive fluids and functional 

 activity within the synapticular region becomes diminished, and it is very doubtful if. as Miss 

 Ogilvie (p. 171 1 suggests, the main advantage is that "an increased endodermal surface is 

 afforded within the visceral cavity." 



a Acting upon this suggestion of Miss Ogilvie, Delage ami Kerouard, in their "Traits de Zoologie Concrete, Tome 

 IT, pt. l', Les Ooelenteres," 1901, have constructed two ingenious diagrammatic figures (pi. 62, figs. 1, 2), attempting 

 p. Bhow how tin- basal infolding of the soft wall of the polyp proceeds in the formation of both liar-like and lamellar 

 synapticula. Tin- polyps of Siderastrxa give no support whatever I'm' such a conception. From the interseptal 

 lamella represented on PI. XXII. tig. 152, it is manifest that each synapticulum is formed independently of the 

 others, not from a continuous infolding of the basal part of the Bkeletogenic layer, as ( Igilvie ami Delage & Herouard 

 assume. 



