Dixon — First Mitosis of Spore-mothev-cells of Lilium. 9 



some a primary chromosome. Elsewtere I have given reasons for 

 regarding these as homologous to the chromosomes of the vegetative 

 mitoses.^ 



"When the chromosomes arrange themselves in the equatorial plate 

 their greatest length is approximately horizontal and radially disposed. 

 Apart from this their orientation seems very various. Thus the Ioojds 

 may lie towards the periphery or towards the centre, and the ends in 

 planes parallel to or inclined to the equatorial plane. In the later 

 stages of the equatorial plate the loops, where they are found, are 

 broken transversely at the bend. In the equatorial plate the chromo- 

 somes become thick and lumpy. The twist of the primary chromo- 

 somes round one another becomes reduced usually to half or quarter a 

 complete turn. The inner end of the chromosome becomes produced 

 into two attenuate processes (fig. 14), which attach themselves to the 

 fibres of the achromatic spindle (fig. 15). This attachment does not, 

 I venture to think, take place as other writers assume, i.e. connecting 

 the primary chromosomes to opposite poles ; but rather each primary 

 chromosome forms connections with fibres from both poles (fig. 15) 

 This view has been forced on me not only by the consideration of the 

 later stages, e.g. during the separation of the daughter chromosomes 

 from the equatorial plate ; but also by actual observation of the attach- 

 ment the chromosomes to the spindle fibres. This observation is best 

 made in sections cutting the equatorial plate obliquely. Then the 

 chromosome may be seen as drawn in fig. 11. Careful focussing of a 

 polar view of the equatorial plate will also sometimes reveal this 

 method of attachment. While an equatorial view of the same, as in 

 fig. 6, will often exhibit a chromosome like that on the right of the 

 figure in which the pointed processes of the chromosome are seen dis- 

 tinctly connecting each primary chromosome to upper and lower 

 fibres. As these attenuate processes have all the appearance of a 

 viscid substance it will be understood that this method of attachment 

 may be effected, no matter how the chromosome is oriented in the 

 equatorial plate, provided that its greatest length is radial. In the 

 case of the more uncommon longer chromosomes which assume the 

 bend in the equatorial plate before alluded to, the attachment taking 

 place in the middle of the chromosome is difficult to make out. But 

 in all probability it is effected in the same manner, i.e. each primary 

 chromosome forms connections with the fibres of both poles. 



1 Chromosomes of Lilium longijlorum. Proc. E.I. A. loc. cit. Cf. Belajeff Ber. d. 

 deutsch. bot. Ges, 1898 ; Ueber EeductionstbeLlen des PflanzeiLkernes. 



