RESEARCHES INTO THE DICYEMIDA. 141 



body. Then the embryo appears as a spherical body, com- 

 posed of a fixed number of cells, amongst which four are 

 remarkable for their unusual shape (fig. 16). The two largest 

 cells form eventually the parietal cells of the urn, and give 

 rise to the capsule, the two middle produce the cover, two 

 cells opposite them the refi-actory bodies, and finally, at a 

 later period, four small cells appear which become the granu- 

 lar body (fig. IT, 18, ^). The other cells of the embryo become 

 ciliated and give birth to the ciliated bofly. The infusori- 

 form embryo then passes into the external medium after 

 traversing the wall of the endodermic cell and the ectoderm ; 

 its mode of life and development after this are not known, but 

 it is a remarkable fact that although the adult Dicyema and 

 the vermiform young are killed by immersion in sea water, 

 still the infusoriform embryo swims uninjured for two, three, 

 and even five days in the same liquid. Apparently, there- 

 fore, the Cephalopoda must be infected by the infusoriform 

 embryo alone. It has not yet been determined whether it is 

 the embryo itself or the ciliated contents of the urn which 

 becomes the Dicyema — whether the embryo passes directly 

 into anewCephalopod or whetherthere is an intermediatehost, 

 and whether or not the infusoriform organism is modified in 

 the sea before it enters the body of its host. There are some 

 other points in relation to the Dicyemidse of which we are 

 yet in ignorance ; thus, we do not know what constitutes the 

 difference between a Nematogenous and a Rhombogenous 

 worm whether an individual, after having produced vermiform 

 young, can in advancing age so modify itself as to bear in- 

 fusoriform embryos ; if reproduction takes places entirely by 

 parthenogenesis, or if the production of embryos of either 

 one kind or the other is preceded by true impregnation. It is 

 just possible that before the infusoriform embryos are pro- 

 duced an ectodermic cell may fuse with the central cell of a 

 vermiform embryo, and if this cell is the male element the 

 fecundation of the endodermic cell is of the same kind as the 

 fecundation of the embryo-sac of phanerogams by the pollen 

 tube, the only diff"erence being that the Dicyemidse would be 

 truly hermaphrodite. 



The chief results of the investigation of the Dicyemidce. 



I. — The Dicyemidae are multicellular organisms formed 

 entirely of cells connected together like the cells of epi- 

 thelium or of vegetable tissue. They have no muscle or 

 nerve-fibres, and in place of internal cavities they have intra- 

 cellular vacuoles. 



II. — All the Dicyemidae are composed of an axial or endo- 



