92 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY. 



hsematosylin. Sublimate preparations with the same stain gave con- 

 firmatory results. In some cases, large spherical pigment granules 

 were found arranged in a circle just outside the disc and close to its 

 boundary. 



Miss Lewis ('96, '98) described the presence of a centrosome and 

 sphere in the giant nerve cells of Clymene producta. She demonstrated 

 the presence of these structures by the use of vom Rath's picric-osmic- 

 acetic-platinic chloride mixture, not followed by a stain. Sublimate 

 fixation followed by Heidenhain's iron-haematoxylin gave confii'matory 

 results. 



The nucleus of these giant cells is excentric and flattened or concave 

 on the side toward the cell centre. The sphere is a more or less sharply 

 defined region lying near the cell centre and has a diameter about one- 

 third that of the cell. It consists of an outer " broad zone of coarsely 

 granular protoplasm ; within this a smaller area of more nearly homoge- 

 neous protoplasm ; and in the centre of this a very small highly refrac- 

 tive body, or occasionally two or three such bodies." The central bodies 

 sometimes have the form of short rods. From the central granule, or 

 granules, ividiations " traverse both the inner, more homogeneous zone 

 and the outer, coarsely gi'anular zone. Sometimes the radiations pass 

 even beyond this into the surrounding, finely granular protoplasm of the 

 cell " ('96, p. 296). The radiations are composed of minute granules 

 arranged in radiating lines. 



Dahlgren ('97) examined spinal-ganglion cells of the dog and found 

 appearances which resembled centrosomes and " centrospheres," but 

 which he believed to be artifacts due to the crystallization of corrosive 

 sublimate. The questionable structure, always found between the 

 pigment mass and the nucleus, appeared in the section as a disc free 

 from granules and having at its centre a black spot. Focussing revealed 

 a radial arrangement of the disc, and the surrounding cytoplasm showed 

 a *' distinct concentric arrangement." 



In some twenty-five cells, before the sublimate had been dissolved out 

 by means of iodine, Dahlgren observed radiating crystals of the subli- 

 mate occupying the region where, in the same cells after staining, the 

 supposed artifacts were seen. 



That the crystals should always be found at the same region of the 

 cell, Dahlgren says, " would indicate some difi"erence in the constitution 

 of the cell at this point." 



Dogiel ('97, p. 108) mentions that, in methylen-blue preparations of 

 the spinal-ganglion cells of certain mammals, " in einer gewissen Ent- 



