jQ HYDROIDA 



which, however, are most frequently concentrated in the jiroboscis, closely to the basis of the tentacles, 

 where the entrance to the proper gastric cave is found. 



Thus there is throughout a typical difference between the Capifata and the Filifera as to 

 endodermal matters, though the Clava apparently presents an intermediate form or a form of depar- 

 ture from which the other types are derivable. Simultaneously the Filifera, as far as can be judged 

 from the data in hand, bear a typical resemblance to the thecaphore hydroids, and here the parallel 

 between the Eiidcndriiim and the CampanulariidcB is particularly obvious. Whether this is owing to 

 a closer affinity or it must be explained only as a phenomenon of convergency, we must at present 

 leave unanswered, because of our deficient knowledge of the group. 



A single family, the Tiihilariida;, shows an anatomic p.eculiarity, as with the species of this family 

 there occurs a peculiar m'esogloeal formation. At the basis of the large tentacles the endoderm has 

 developed a thick supporting cushion formed as a ring of mesogloeal tissue of large cells round the polyp. 

 This leans inwardly on tlie supporting lamella, and is bounded against the axial endoderm of the tentacles 

 by a delicate membrane, which in some cases it is rather difficult to point out. 



Hollow tentacles occur in two ways. In their original .shape they are, as is the case with 

 Hydra, openly communicating with the gastric cave of the polyp. This state of things, however, 

 has ceased with most hydroids and cannot be found in any of our northern species. In these, on the 

 other hand, occasionally occurs, as in Clava iinilticorms and in Myriotlicla a central cavity in 

 the tentacles, at any rate in their basal part. This cavity, however, does not communicate with the 

 gastric cave of the polyp, but is basally bounded by the unbroken supporting lamella. This central 

 cavity of the tentacles, as it is represented by well-developed Clava multicornis, might be looked upon 

 as a primitive condition of things. However, in forms so highly organised as Myriothela, it must 

 sooner be considered as a secondary phenomenon, which cannot have any direct correspondence with 

 the primitive condition of things in Hydra. 



In Myriothcla the tentacles show a peculiar structure, elsewhere unknown in the hydroids. 

 The matter is more precisely described by Jaderholm (1905). The .supporting lamella is in the 

 thickened distal part of the tentacle transformed into a cushion constructed by delicate radially 

 placed fibres, showing no cellular structure and densely crowded. They seem to be intended for 

 strongly stiffening the distal portion of the tentacles and for making the armed outmost portion of 

 the tentacles larger and more powerful of resistance. 



On the ground of the anatomical features stated, and of morphological characters hitherto turned 

 to account in systematics, is brought about a system which, as far as the athecate hydroid families 

 are concerned, can be suunued up in the following key of determination: 



A. No formation of gonophores. Eggs and sperms, developed in the wall of the polyp. The tentacles 

 — if such ones occur — hollow, openly communicating with the gastric cave (Sectio Simplicia nov.) 

 Fam. Hydrida. 



B. The generative cells developed in special gonophores. 



I. The polyps with large, broadly oval or spherical ncniatocysts (vSectio Capitata Kiihn). 

 a. The tentacles of the polyp wholly or in part claviform. 



