4 AMERICAN HYDKOIDS. 



viiies. In some cases, o. g., Campanularia sjieciosa, thej^ appear as if flattened out on the ob- 

 ject over which they grow, thus increasing their adhesive surface and securing a correspondingly 

 firm anchorage. 



Some wiitci-s have contended that the stolon or rootstock is really a recumbent stem from 

 wliich the pedicels arise, being confined to the uppermost side by the necessities of the case. 

 In regard to such suggestions it seems to me that it is unnecessary to discuss these homologies 

 from the fact that the various parts of the hydroid colony, branches, stem, stolon, hydrorliiza, 

 etc., ai-e really convertible terms. As I have said before : 



As a matter of fact, the subject of homology among Hydroida has been unnecessarily obscure because the whole 

 group is so primitive that any one part is homologous with several others, or rather that the parts are not greatly or 

 fundamentally differentiated. For instance, the creeping rootstock may properly be regarded as a portion of the 

 hydrorhiza in many species; in many others it is considered a true stem, or hydrocaulus, which has adopted a pro- 

 cumbent habit; again, as we have just seen, it takes the guise of an accessory tube in a fascicled stem, which may in 

 its distal portion adopt the further disguise of a branch or even a hydrocladium.' 



The stem. — ^As already stated, the stem in most campanularians is monosiphonic. This 

 is the situation in the numerous forms in which the creeping rootstock is regarded as a pro- 

 cumbent stem as well as in all cases where the stem is erect, with the exception of a few species 

 discussed below. When erect and simple the stem is ordinarily divided into regular intemodes 

 with a group of annulations just above the origin of each branch or pedicel. In Campanularia 

 flexuosa (fig. 5) the pedicel itself appears to be a curved upward extension of the internode be- 

 low, producing an effect as if the stem itself were made up of a series of long pedicels curved 

 alternately to the right and left. 



In Obelia geniculata (fig. 4) the internodes are remarkably short and stout and much broad- 

 ened at their distal ends so as to form a conspicuous shoulder upoii which the pedicel appears 

 to be emplanted; the perisarc being greatly thickened on the outer side of the stem below the 

 insertion of the pedicel. 



The stem is fascicled in but a few Campanularidee. Campanularia verticiUata (fig. 7) 

 shows a unique condition in that the compound stem is made up of a bundle of closely adher- 

 ent and almost exactly parallel tubes, each of which gives off a series of pedicels, these being 

 so spaced that verticils of pedicels are produced at fairly regular intervals. Hargitt^ says 

 that he had found coenosarcal connections between the tubes. I have been unable to satisfy 

 m)'self of such connections; but, as I did not have fresh material to study, and as I well know 

 Doctor Hargitt's scrupulous care in making and recording his observations, the facts are doubt- 

 less as stated by him. 



The compound stems of the Campanularidse differ from those of the Plumularidas and 

 Sertularidse in being destitute of the "axial tube" always found in polysiphonic stems in these 

 families.^ Any of the component tubes may bear branches or pedicels in the Campanularidse 

 while only the axial tube bears these in the other families mentioned. 



In examining a portion of the stem of C. verticiUata in which the component tubes have 

 been separated by boihng in a potash solution it is seen that a new tube may arise at almost 

 any point by budding from an old tube. It seems as if in such cases the origin of a new tube 

 were homologous with the origin of a pedicel; in the former case the zi&n process becoming 

 another tube adhering to the stem, and in the latter it remains free and terminated in a 

 hydrotheca. 



In Clytia universitatis is found another stem fascicled in the manner just described, and 

 in this the writer was able to demonstrate coenosarcal connections between the component 

 tubes greatly resembling those described by Hargitt. Hartlaub ' has made a very careful study 

 of the compound stem of Obelaria gelatinosa (fig. 6) and has illustrated his work with admirable 

 drawings. He fuids that this stem is made compound by the aggregation of a number of sto- 



' Part I, p. 7. 



2 A few Cadenterates of Woods Hole, Biological Bulletin, vol. 14, No. 2, p. 113, and fig. 17. 



' See Part I, pp. 4-8, and Part II, pp. 5-7. 



* Die Ilydromedusen Helgolands, 1897, pp. 488^95, pi. 17. 



