VALUE OF THE ATTRACTION-SPHERE. 327 



genital cells of the salamander. The chromosomes are 

 attached to the fibres of the radial envelope, while the 

 Verbindungsfaden (i.e., the fibres in the equatorial region 

 of the spindle after the chromosomes have separated) are 

 produced, according to Ischikawa, by strands of linin 

 becoming stretched out between the separating chromo- 

 somes (Fig. VI.). 



The sphere is in Noctiluca singularly well adapted for 

 the exhibition of the concentric zoning of its structure, even 

 its external envelope being sometimes figured by Ischikawa 

 with a sharp outside boundary, beyond which there stretch 

 the less definite protoplasmic contents of the cystoflagellate's 

 body. I have myself observed in the embryonic genital 

 cells of the salamander a similar sharp boundary to the 

 outer surface of the radial envelope of the sphere, and 

 when we bear in mind that the resting centrosomes, as 

 Heidenhain insists again and again, are to be regarded as 

 the middle point of the whole cell, it is quite possible that 

 the concentric zoning may have some relation to osmotic 

 influences operating on the inner structures from without. 



However this may be, the arrangement of the parts of 

 the sphere in diatoms is strangely different from the above. 

 As described by Lauterborn (24) the sphere in these 

 monoplastids is represented at the commencement of mitosis 

 by a small globe and a rod, which lie outside the nucleus 

 and are quite separate from one another. The globe is the 

 centrosome, and from it stretch radiations comparable to 

 those of the radial envelope of a compound, or the single 

 radial zone of a simple sphere. The rod swells up into 

 a spindle, round which the chromosomes become normally 

 grouped, while the centrosome, dividing, passes to each end 

 of the spindle figure. It will thus be seen that the sphere 

 of a diatom is in the same dismembered condition as the 

 sphere I have described in the spermatocyte of a rat, and 

 that thus the monoplastid analogue of the most curious 

 modification in the polyplastid sphere which I have come 

 across is found among the diatom, one of the most 

 aberrant forms of all unicellular organisms. 



Up to the present time, the whole cell has been usually 



