144 Crozv.—Dimorphococcus Fritschii , a New Colonial 
Dictyosphaerium and Sciadium , yet the details of structure of the latter 
show it to be a genus of the Heterokontae, and the resemblance to Dictyo¬ 
sphaerium to rest on analogy only. In view of the differences in details of 
structure mentioned above, Bohlin thinks that the same applies to the sup¬ 
posed close relation of Dimorphococcus and Dictyosphaerium , although both, 
of course, must still be classed in the Isokontae. 
The structure of the partial colony Dimorphococcus Fritschii , n. sp., and 
probably also, to a lesser extent, that of D. cordatus , Wolle, helps to bridge 
the gap between the genera Dimorphococcus and Dictyosphaerium. The 
fact that each cell possesses a distinct stalk, together with the character of 
the stalks themselves, as well as the presence of a colonial mucilage envelope 
in the new species, all recall the morphological features of Dictyosphaerium. 
Finally the presence of the small lappet at the base of each cell, if our inter¬ 
pretation of it is correct, shows that partitioning of the mother-cell wall into 
four parts is not unrepresented in Dimorphococcus , although the greater part 
of each segment may appear as a mucilage thong at a very early stage. 
Bohlin thinks that the partial colonies of Dimorphococcus are homologous 
with the colonies of Scenedesmus. This is strongly supported by his descrip¬ 
tion of D. lunatus , A. Br. In the latter the cells do not possess individual 
stalks, the very short strands sometimes connecting adjacent cells being 
comparable with the intercellular pads of some species of Scenedesmus. 
The four cells of the partial colony, generally closely adpressed (see e. g. 4 , 
PI. CXXI, Fig. 5), do not develop their corresponding mucilage thongs 
until they divide. For until the cells of the partial colony divide, they are 
enclosed in a gelatinous mass formed from the parent wall but not yet 
organized into thongs. Thus in comparison with Dimorphococcus Fritschii, 
n. sp., and the species of Dictyosphaerium , the mucilage threads of Dimorpho¬ 
coccus lunatus , A. Br., are late in development to the extent of one cell- 
generation. 
Dimorphococcus Fritschii , n. sp., appears to be closely allied to D. cor¬ 
datus , Wolle. 
The latter is not well known, and the figures of Chodat ( 2 ) differ 
remarkably from those of Wolle (6, PL CLX, Figs. 30-38). It would 
appear that D. cordatus , Wolle, generally has smaller colonies than D. 
Fritschii , n. sp., is somewhat less regular in colony form, shows much less 
marked dimorphism in its cells, and above all lacks the colonial mucilage 
of our species. 
In conclusion we give a diagnosis of the new organism and a sum¬ 
mary : 
Dimorphococcus Fritschii , n. sp. 
Coenobio magno, ellipsoido, regulario, denso, coenobiis secondaries 
quaternarum cellularum compositis uniquoque certa vagina gelatinosa cir- 
