314 Proceedings of the Royal Irish Academy, 



archaic tendency of cells to radial differentiation, such, as reaches its 

 maximum in the Eadiolaria. 



The component rays of these stars are themselves usually simple 

 rods ; but, in some cases, they certainly appear to be in course of 

 bifurcation. 



An appearance of such bifurcation might, it is true, be produced 

 if one ray were lying over another in its proximal portion ; but in 

 several instances it seems undoubtedly to be a case of real forking. 



This raises the question as to the plane in which the rays lie — a 

 question by no means easy to decide. The specimens being very 

 deeply stained and mounted in balsam, it is difficult to be absolutely 

 certain on this point ; but in some cases there is sufficient difference 

 in focus to suggest that they radiate in three dimensions. The speci- 

 mens were dried on the cover-glass in the usual method of mounting, 

 and, under such circumstances, the rays even of a " solid " star would, 

 of course, tend to flatten on its surface. 



In a few cases there is a transverse constriction of some of the 

 rays, indicating that the organisms may multiply by abstriction of the 

 distal portions (PI. VII. fig. 15). 



JS'o spores have been observed. Though in some examples single 

 highly -refracting spots are observable near the tip of each ray, these 

 are probably due to plasmolysis. 



The remaining feature of importance to which I would call atten- 

 tion is the presence, in a large number of specimens, of a central 

 colourless spot. This seems constant in the older, multiradiate types, 

 but does not occur in the simpler forms. "With a high magnification 

 it will be observed, in some cases, that the bases of the rays are 

 rounded off and project somewhat into the light area, while occasional 

 specimens may be seen in which the central space communicates with 

 the exterior (PI. VII. figs. 16 and 17). This condition suggests that, 

 in course of time, the stars disintegrate by the absorption of the 

 central protoplasm into the several rays which then separate, and 

 become the rod-like bacteria with which we started. 



The foregoing description includes all I have been able to observe in 

 the preparations, and the only explanation thereof which seems to me 

 tenable. These numerous stellate organisms cannot be chance aggre- 

 gates of rods ; and it is equally difficult to imagine the development 

 of a successive series of rays from a single point whether a spore or 

 the middle of a rod-like form. No other unicellular organism, 

 so far as I know, affords any evidence in favour of such a view, and 



