294 " ENDEAVOUR " SCIENTIFIC RESULTS. 



and Halopteris carinata, AUman, in the Halopteris group. 

 For similar reasons, but little significance can be attached to 

 the polysiphonic or monosiphonic condition of the hydro - 

 caulus. The presence of " intermediate internodes " in the 

 hydrocladia is not even of specific importance in such species 

 as P. campanula, where these internodes may be distinctly 

 defined in parts of the polypidom, while in other portions the 

 nodes dividing them from the hydrothecal internodes may be 

 obsolete. But the presence of intermediate internodes in 

 P. setacea and its allies, and their absence in such species as 

 P. hadia, are more constant characters. 



On the whole, I am disposed to think that the best primary 

 sectional division of the genus (setting aside Antennella and 

 Monotheca) would be into two groups, one containing those 

 forms in which hydrothecse are borne on the rachis as well as 

 on the pinnse, the other for all the species in which no hydro- 

 thecse are borne on the rachis. This group, in which the genus 

 reaches its highest development, includes the minor groups 

 represented respectively by the well-known P. setacea, P. badia 

 and the aberrant P. asymmetrica. The other main group, of 

 which P. catharina and P. campanula are typical, I would 

 distinguish by the sectional name Thecocaulus. It is the 

 more primitive type, being the next stage of development 

 from the unbranched Anteyinella. I may here refer to a 

 question of terminology, namely the use of the term " hydro- 

 cladium." Billard includes in the genus Antennella " les 

 especes de Plumulariidae reduites aux hydroclades, se 

 detachant de I'hydrorhize, ou meme de la base d'un hydro- 

 clade primaire," while Jaderholm says that in Antennella 

 there are no hydrocladia, but only hydrotheca-bearing stems. 

 Billard's expression might be held to imply that Antennella is 

 a degenerated form of Plumularia, while Jaderholm evidently 

 regards it as a primitive form which has not developed 

 branches. Of course it might be urged that any part of the 

 polypary which bears hydrothecse is a hydrocladium, but this 

 would imply that in a pinnate specimen of P. campanula, for 

 example, the pinnse, the branches, and even the polysiphonic 

 stem, were all hydrocladia, obviously a reductio ad absurdum. 

 I think, therefore, that in this species the simple shoots, 

 {P. indivisa) should be described as stems, while the lateral 

 branches, whether scattered or regularly pinnate, are 

 hydrocladia, which in their turn may bear secondary hydro- 

 cladia. According to Billard the true Antennella is a species 

 which does not exist in the pinnate form, and I think that 

 in the majority of species at least, it is a primitive, not a 

 degenerated form. Nevertheless I am acquainted with three 



