164 Dr. C. Chun on the Siphonophora. 



the Diphyidae with two, and the Polyphyidae with more 

 tlian two nectocalyces. The family for which I propose the 

 denomination Polyphyidae, however, shows several peculiari- 

 ties, as the most noticeable of which it is to be indicated that 

 the individuals are certainly distributed in clusters upon the 

 stem, but that they do not become free in the form of Eudoxice. 

 Male and female nectocalyces possess a remarkably small 

 umbrella, and bring the sexual products to full maturity in 

 the large manubrium, without separating from the stem as 

 Medusae. While in the Monophyidas and Diphyidse the cyclical 

 development is distributed over two or three generations, 

 these are here compressed into one. 



But how is it to be explained that a direct development 

 prevails among the Polyphyidae and Physophoridae, to give 

 place again to an alternation of generations in the most highly 

 organized Siphonophora, namely the Velellidae? In order to 

 answer this question we must go a little further afield. As I 

 have already indicated, the Calycophoridae possess a hydro- 

 static apparatus in the form of the so-called fluid-receptacle 

 with its oil- drop. Now in all other Siphonophora, in place 

 of the specifically light oil, a compressible medium, a gaseous 

 mixture, is secreted at the upper extremity of the stem. In 

 them a new organ, the air-sac [pneumatophore], makes 

 its appearance j and this, which is originally of small size, 

 gradually acquires more considerable dimensions, until, 

 in the Bhtzophysce, Physah'ce, and Velellce, it fundamentally 

 influences not only the physiognomy, but even the whole 

 organization. As regards the development of the air-sac, I 

 can confirm MetschnikofPs statements from his investigations 

 on the embryos of Halistemma pictum ( = JL tergestinum, 

 Claus). At the pole of the planula, which advances foremost 

 in locomotion, we observe a solid thickening of the ectoderm, 

 which is finally constricted off from its origin, and, surrounded 

 by the small-celled endoderm, passes somewhat inwards. By 

 the separation of the constricted ectodermal cells a cavity is 

 produced, which is filled with granular fluid and rapidly 

 dilates. The ectodermal cells, with the exception of the por- 

 tion turned towards the hinder pole of the planula, secrete a 

 delicate chitinous lamella towards the inner cavity of the 

 vesicle, and at the same time begin to secrete a gaseous 

 mixture, which accumulates above the fluid. The perfectly 

 closed air-sac early acquires a flask-like shape, and in the 

 true Physophoridae never communicates with the outer world. 

 In its relatively inconsiderable development it plays only a 

 subordinate part, so far as locomotion, %. e. a rising and sinking, 

 is concerned. On the other hand, this is very effectively per- 



■ 



