HTDROIDA. BALE. 27 



Billard ranks 8. divarieata as a synonym of S. johnstoni, 

 but as he does not include 8. subdichotoma also among the 

 synonyms, he evidently considers it a distinct species. Assuming 

 such to be the case, the inclusion of 8. divarieata would be 

 untenable, since the most intimate relationship of 8. divarieata 

 is with 8. subdichotoma, not with 8. johnstoni. 



The accurate and detailed description by Hartlaub leaves 

 little room for further remarks on the structure of the species. 

 The principal point on which my observations do not agree 

 precisely with those of Hartlaub is in regard to the jointing of 

 the hydrocaulus, I have not found, either in 8. divarieata or 

 8. johnstoni that the normal condition is for the internodes to 

 support several hydrotheca?. Setting aside the question as to 

 the original condition of the internodes — to which I have referred 

 on a former page — I have found in the polypidoms of S. johnstoni 

 and of most of the varieties here ranked under 8. divarieata 

 that the typical condition of the genus — one hydrotheca to each 

 internode — is the rule, and the opposite condition quite 

 exceptional. In 8. divarieata, type, longer internodes are more 

 frequent, and in var. mbdicltotoma they preponderate, so that 

 often the only parts where single-celled internodes are to be found 

 are the latest-formed distal portions. Here the perisarc is 

 extremely thin and delicate, contrasting strongly with the much 

 thicker portions which are but little older, and it can easily be 

 understood how readily the faintly-marked nodes may disappear 

 in the process of growth. There are no internal teeth in the 

 hydrotheca, but the apocauline wall is often thickened at the 

 margin between the lateral teeth, which gives it when seen in 

 side view almost the aspect of an internal marginal tooth. The 

 oblique internal septum is not comparable to the permanent 

 intrathecal ridge of 8. crenata or the Plumularians, but seems to 

 be an almost membranous film of extreme delicacy, easily 

 detached, and leaving a slight mark on the hydrotheca wall after 

 its disappearance. 



The gonangia are more complicated structures than those of the 

 rugosa-gronp, where the annular ridges are simple undulations of 

 the gonangium-wall. Here they are more or less thickened, 

 forming a ring of a cup-like form, from within the margin of 

 which rises the next undulation, while the ring is extended m the 

 form of a narrow flounce-like film, projecting upward. Just 

 below this the perisarc is generally marked with delicate wavy 

 stria:, or sometimes faint reticulations. At the summit of the 

 gonangium, which leans forward away from the hydrocaulus the 

 ridges are completely annular, but on the lower part they are 

 wanting at the back. After passing round the sides of the 

 gonangium they generally take an upward course and gradually 



