96 



MEDUSA. I. 



southerly parts of its area of distribution. It is possible, accordingly, tliat the species just described 

 will prove in time to be only a northern giant variety of a species already known to science; but as 

 long as it cannot be referred with certainty to any known species, I prefer to describe it as an inde- 

 pendent species, for which I propose the name of PhialhiiiDii islandiciuii, because all the specimens in 



hand have been found in the neighbourhood of Iceland. 



Chart XII. Finds of Phialidium islandiczim nov. sp. 



The species is quite distinct from Phialidiiiiu /icmisp/icrricuin, not onh- b)- its size and the 

 large number of tentacles, but also by the number of marginal vesicles never exceeding the number 

 of tentacles. Furthermore the mouth-lips are larger and more crenulated than in the case of Phialidmiit 

 hemisph(F.riaim^ and the gonads are longer. The sagittal sections through the tentacular bulbs show 

 (Plate V, figs. 2 and 3) that the ectoderm on the adaxial side of the bulb is more highly developed 

 in Phialidium hemisphcrricum than in Ph. islandicinn. The trace of an abaxial process on the base of 

 the tentacular bulbs in Pliialidium is/audicniii is not always distinct; in Ph. hemisphceriaim it is entirely 

 lacking. The shape of the marginal vesicles (see the sections, Plate IV, figs. 13 and 14) presents no 



