ROBERTSON: THE CHROMOSOME COMPLEX. 277 



SYRBULA ADMIRABILIS. 



do not always lie in a plane parallel with the field of the ob- 

 jective. The image is thus foreshortened. The proper size 

 relations, however, are quite apparent to the eye of the ob- 

 server as he examines the chromosomes by different focusings. 



In mitosis every chromosome divides longitudinally. The 

 splitting seems to have already taken place when the elements 

 first appear in the equatorial plate. (Figs. 1-4.) The separa- 

 tion of the halves begins at that end of the chromosome which 

 is turned towards the center of the plate (fig, 6), the end to 

 which the fibers from each pole of the spindle are attached. 

 It seems sometimes that the halves are moving apart at the 

 outer end of the chromosome but this only indicates that the 

 element is already split. The real separation of the halves be- 

 gins at the inner end. 



At this point it may be well to call attention to the position 

 of the members of a pair in the plate. The two may lie to- 

 gether but it does not follow that they always do so. In the 

 case of the smaller chromosomes the members of a pair are 

 quite often on entirely opposite sides of the plate, or on the 

 same side but with several other chromosomes between them. 

 Compare figure 6, chromosomes 4, 4 and 3, 3 ; and figure 3, 

 chromosomes 3, 3 and 2, 2. In figure 6, No. 4 on the right 

 side has no member near it that at all corresponds in size, and 

 yet we know that it will unite only with one which is its equal. 

 We know this from the fact that the halves of a tetrad are 

 always alike in size, length and diameter. The only chromo- 

 somes with which this one can at all conjugate are on the 

 opposite side of the plate. 



THE ACCESSORY CHROMOSOME AND THE NUCLEOLI. 



In studying the resting stages of the spermatogonial and 

 spermatocyte nuclei the chief object has been to trace the ac- 

 cessory, to see if there is any connection between it and the 

 nucleolar-like structures that are always present in these 

 stages of the nucleus. The earliest of the secondary sperma- 

 togonial nuclei that were found (figs. 7, 8 and 9) showed the 

 presence of these nucleolar structures. With Flemming's tri- 

 color they take safranin, like chromosomes in the condensed 

 or homogeneous condition (figs. 7-10). Distinct from the 

 nucleoli the accessory chromosome lies at one side of the nu- 

 cleus, separated from the rest of the chromosomes in a sac or 

 vesicle of its own (figs. 7-11). It shows no evidence of being 



