HYDROIDA II 



hydrotheca, which is somewhat thinner than the wall of the hydrotheca generally, and can be brought 

 in over the aperture of the hydrotheca, so as to close it. Within the family of Campanulinidtz, several 

 types of lid have been developed, some consisting of the distal part of the hydrotheca (subfamily 

 Cuspidellmce) others formed by the primary roof of the hydrotheca (subfamily Calycellince). The 

 closing apparatus in the first case consists of folding, membranous parts, roof-shaped or conical, covering 

 the opening of the hydrotheca. In the latter case they form a conical roof, consisting partly of folding 

 membrane, partly of triangular plates; in this group, the closing apparatus is always sharply marked 

 off from the hydrotheca, whereas in the Cuspid ellina, it passes gradually over into the same. In Ser- 

 tulariida also, we find several forms of closing apparatus either consisting of some few separate tri- 

 angular plates, and thus exhibiting the same type as certain Campanulinidce, or reduced to two - 

 sometimes even a single plate — but they always appear to be formed from the original roof of the 

 hydrotheca, in contrast to Toichopoma and CuspidellincE. 



These features have been the subject of general attention in systematic respects; less considera- 

 tion, on the other hand, has been devoted to the different structural conditions in the polyps them- 

 selves, as arising from varying conditions in the endoderm. 



In Lafoeida and Campanulinid<z, the endoderm is, as a whole, but little differentiated; here 

 also, however, we find an oral part above the tentacle whorl, where the indifferent cells with small 

 nuclei are decidedly in the majority, and the structure of the polyps here strikingly resembles that of 

 the Bougainvilliidm. In Haleciidce, on the other hand, we find practically throughout, a marked distinc- 

 tion below the tentacle whorl between a fore-stomach and the actually digesting, basal part, the sto- 

 mach itself. While the digestive cells in the former are decidedly in the minority, it is they alone 

 which form the endoderm of the stomach. The same differentiation into two stomach sections is still 

 more pronounced in PlumulariidcE, and in the polyps of both families we must notice a ring-shaped, 

 external constriction of the polyp, where the boundary between the two endodermal zones must be 

 taken to lie. The same division of the endoderm of the polyp into two regions is likewise found in Agla- 

 opheniidm, but it is not always so easily distinguishable here on mere external observation of the 

 polyp, being in particular frequently obscured by the ectodermal extensions previously referred to, by 

 which the hydranth is attached to the inner ribs of the hydrotheca. 



• Even more pronounced is the differentiation of the endoderm in most of the Syntheciida and 

 Scr hilar iida>. Here, the polyps are as a rule furnished with a single (ventral) blind sack placed 

 abcaulinally to the colony. The endoderm of the stomach parts, otherwise formed by a high cylindric 

 epithelium, forms in the blind sack only a low epithelium, which does not appear to contain digestive 

 elements. This seems to suggest that Kiihn's supposition (1913 p. 68) is correct, and that the abcaul- 

 ine blind sack in these families serves for storage of nutritive matter to be digested subsequently, when 

 the polyp is again in a resting position. The abcauline blind sack must be said to be characteristic of 

 the two families, even though in primitive forms such as for instance Thyroscyphus and Dynamrna, 

 only an indication of it may be found. 



Finally, there is yet another type of polyp found in Campanulariida and Siliculariidce. While 

 all the remaining families of thecaphores are characterised by a conical peristome, this is, in the 



two mentioned families, clubshaped, situate with a narrow base above the tentacle whorl on the broad 



1* 



