114 bulletin: museum of comparative zoology. 



Very often these granules, which occur along the course of a radiation, 

 mark the point of divergence of secondary fibrils. In Figure 1, if the 

 radiation which is most nearly in the axis of the cell be followed for 

 one-third its length from the centre, a granule will be seen from which 

 two fibrils proceed on toward the periphery of the cell. Figure 2 shows 

 similar conditions, but the regenerated cells (Plate 2, Figures 10-13) 

 exhibit this arrangement more clearly. 



Figure 2 represents a cell of the same type as that of Figure 1 ; but 

 the central body is much nearer the nucleus, and is either elongated or 

 else consists of two granules very close together. The central clear 

 space is confluent with the clear spaces around two granules that lie in 

 the courses of radiations from the central body. 



Figures 5 and 6 are from acetic-sublimate preparations stained in iron- 

 haematosylin. They represent cells very similar to those just described. 

 The attempt has been made to show something of the difference be- 

 tween the appearance of an osmic-acid and a sublimate preparation. 

 The nuclear structures are fairly well brought out by the sublimate, but 

 not so clearly as by the osmic acid. The structure of the cytoplasm is 

 much less definite, but as far as it could be made out, it agrees with the 

 description given for osmic preparations. Both cells show some evi- 

 dence of an arrangement in the cytoplasm concentric about the centre 

 of the radiating system. This is especially well marked in Figure 6. 



Cells of the Second Type. — Figure 3 represents a cell in which a 

 further complication of the radial system is introduced. The central 

 body in this case appeared very large and of irregular form, but careful 

 focussing resolved it into three granules of different sizes. They are 

 surrounded, not by a clear space, but by a region which appears dense 

 and very finely granular. The limit of this region is emphasized, in 

 the section, by the occurrence of a number of stained granules which 

 are arranged in the arc of a circle about the central granules. The 

 radiations from the centre can be seen to include in their courses cer- 

 tain conspicuous granules lying in this arc. This finely granular region 

 is evidently comparable to the "sphere" described by many authors. 

 Figure 3 is strikingly similar to some of the figures in Bilhler's paper 

 of 1895. 



Figures 7, 8, and 9 (Plate 2) represent small cells of this type from 

 the brain. In Figure 7 the body of the sphere differs in no visible way 

 from the surrounding cytoplasm, but a number of granules form a 

 complete circle about the central granule. In Figure 8 the entire radiat- 

 ing system is very faint, but there are evidences of two concentric cir- 



