MATTHAI— RECENT COLONIAL ASTR^ID^ 7 



Gardiner too could hardly distinguish cell outlines in the ectoderm and endoderm in his 

 sections of Coenopsammia and Flahellum ; with regard to the calicoblastic layer of the 

 latter genus he says (50, p. 139), " I can only regard the layer as an enormous syncytium, 

 and for the growth of a septum there would seem to be a flowing up of the protoplasm on 

 either side." The only structures with definite boundaries which may be regarded as cells 

 are, as suggested below, the nematocysts and reproductive elements. Whether the 

 syncytial appearance of the body-layers is a secondary feature due to the fusion of cells or 

 is the original condition persisting in the adult cannot at present be decided. It is possible 

 that the two layers may arise and continue as sheets of protoplasm wuth the nuclei 

 multiplying by division without corresponding fission of the protoplasm. But of course no 

 definite view as to the real nature of these layers can be formed without studying their 

 development and without fresh polyp-tissue. 



Structures similar to the gland cells (both mucous and granular) of Actinians as 

 described by various authors are present in the ectoderm. In my sections they look like 

 vacuoles, their cellular nature being doubtful. They often appear non-nucleated or with 

 more than one nucleus in their walls. The mucous type (figs. 2 and 3) is oval or flask- 

 shaped, usually with the broader end turned towards the free surface ; it occurs in varying 

 numbers in all the species, in the oral-disc, edge-zone, tentacles and the convolutions of 

 the mesenterial filaments, and contains some kind of mucous material which stains brown. 

 The granular type (figs. 11 and 66) is either flask-shaped or elongated, the granular 

 contents, whose nature and function are uncertain, staining dark ; it is usually present in 

 the mesenterial filaments, less frequently elsewhere in the ectoderm. 



The Hertwigs (66) supposed the granular cells to be stages in the development of the 

 mucous ones, whereas H. V. Wilson (122) regarded the two types as distinct, since they 

 are distinguishable from the start and have a different distribution in the early develop- 

 mental stages of Manicina areolata. 



Small oval sacs, showing finely granular contents stained light pink, are usually 

 present in the ectodermal ridges of the stomodaeum as well as in the mesenterial filaments 

 (figs. 23 and 45). These perhaps belong to the granular type and their contents may be 

 digestive in function. 



The endoderm is often vacuolated, the vacuoles being oval or goblet-shaped ; nuclei 

 are less numerous than in the ectoderm and occur between the vacuoles, or the latter may 

 appear multi-nucleated. The vacuoles are usually more or less transparent, except in the 

 mesenteries, where in some species, e.g. Cyphastrea serailia (Forsk.), they stain' reddish- 

 brown (fig. 4). The exact nature of these vacuoles is unknown. 



Willem (120) regarded the endoderm as mainly absorptive and excretory and the 

 mesenterial filaments as capable of secreting a digestive fluid. In addition to extra- 

 cellular digestion Willem also refers to intra-cellular digestion in the endoderm, whereas 

 Gardiner (49, p. 375) rejected the possibility of the endoderm having any digestive 

 function, regarding the layer as being homologous with the mesoderm of Triploblastica. 

 Bourne (17, p. 225), however, later suggested that in Heterocyathus cequicostatus the 

 "glandular-looking cells" may secrete a digestive fluid, and thus disagrees with 

 Gardiner. 



