iqi6] NOTHNACEL— REDUCTION DIVISIONS 469 



The bivalents are now open at both ends and not quite as large 

 as formerly, decreasing from now on (fig. 24). When they first 

 loosen up, fibers are not fastened to each chromosome apparently; 

 but as the multipolar complex forms, a weft becomes moored to 

 each one, in many instances so conspicuous (figs. 29-35) that it 

 appears to be homogeneous structure at the point of attachment, 

 owing to the thick fibers and also to their close association. 



Each chromosome, as stated by Lawson (20), is saturated with 

 karyolymph and is an osmotic system; although this does not 

 necessarily mean that the old membrane must surround each 

 member in order to accomplish this. If such were the case, the 

 membrane would have to break up into the proper number of 

 pieces, wrap about each chromosome, and then become sealed; 

 a process which is far more complex than ever before attributed to 

 a nucleus. Each chromosome will have the power to develop its 

 own membrane, owing to diffusion of the sap from it, and since 

 each then will have the same osmotic power as the nucleus did as 

 a whole, the same process will continue as it did with the nucleus, 

 resulting in the formation of a weft of fibers from each individual. 

 The fact that these wefts are distinct from the other fibers points 

 strongly toward this idea. 



Lawson's theory that the shift from multipolar spindle to 

 bipolar spindle is an expression of "lines of tensions" appears to 

 the author to have little or no foundation, since the reticulate 

 cytoplasm is not forced to occupy less cubical space than formerly. 



The paired chromosomes, which formerly were tightly twisted 

 about each other, gradually come to lie upon the equatorial plate; 

 the fibers formed outside the nuclear membrane extending from 

 pole to pole and the individual tufts of fibers attached to the chro- 

 mosomes extending to but one pole. At metaphase the separation 

 of whole somatic chromosomes, that previously were end to end 

 in the spirem, is completed, this reduction being immediately 

 followed by the longitudinal separation of the halves (fig. 35) of 

 each of these, the origin of which was seen in the resting nucleus 

 (figs. 2-10). Those believing in the pairing of somatic chromo- 

 somes or spirems in early prophase or pseudo-reduction have little 

 to say concerning the origin of this split; while those claiming that 



