47 2 Farmer. — On Spore- Formation and 
these become very faint for a time, a circumstance often met 
with in spindles when at this stage in their development. 
Presently, however, the radiations again become clear, and the 
daughter-chromosomes move off to their respective poles. 
There is a sheaf of fibres commonly attached to each daughter- 
chromosome, and the whole process of the separation of the 
daughter-chromosomes is such as to suggest a passive ‘ roping- 
up 5 to the poles, rather than any spontaneous movement on 
the part of the chromosomes themselves. 
It is easy to distinguish the number of the chromosomes in 
the archesporial nuclei. They are, in all the nuclei in which 
I could count them, sixteen in number. I estimated them 
both at the equatorial plate-stage and also, in other instances, 
while the daughter-chromosomes were travelling polewards. 
The membrane very early appears around the daughter-nuclei, 
but the connecting spindle-fibres persist for a considerable 
time, and across them the cell-plate is formed in the normal 
manner. I was unable to distinguish the centrospheres of the 
two daughter-nuclei after their walls had been formed. I saw, 
occasionally, vesicular bodies which occupied the place where 
the centrosphere might have been expected to be found, but 
they were very irregular in character. Sometimes they were 
altogether absent, and at others I saw as many as three lying 
together. The protoplasm is somewhat vacuolated at this 
stage, and I believe that the structures in question were 
nothing more than small vacuoles which happened to lie upon 
the surface of the nuclei. The nucleoli reappear early within 
the nuclei, first as two or three small bodies which finally fuse 
to one large one, and the linin concomitantly loses its 
chromatin constituent, and passes quickly into the ordinary 
resting condition. I was unable, however, to distinguish any 
difference in the staining capacity of either the nuclear sap or 
of the cytoplasm, such as would indicate that the stainable 
substance (chromatin) had become diffused out of the linin. 
The only structures which do change in this respect are the 
growing nucleoli. This does not necessarily imply that the 
chromatin passes, as such, into the nucleolus. Indeed, all the 
