226 



REPRODUCTION 



split chromosome as one, there are only half as many here as 

 appeared in the prophases of the grandmother cell. The divi- 

 sion of the mother-cell nuclei' which now ensues consists es- 

 sentially in the separation of the longitudinal halves of the split 

 chromosomes and their distribution to opposite poles, where 

 they form the nuclei of the microspores, four descended from 

 each grandmother cell (Fig. 128, 13, 14, 15 and 16). The micro- 

 spores have just half as many chromosomes as did the grand- 



FiG. 130. — From nuclev of grandmother cells of pollen grains of Funlcia Sieboldiana 

 A, in diakinesis showing pairs of chromosomes of different sizes; B, metaphase, the members 

 of each pair separating as shown by widely diveiging ends of r and s\ r and 5 each show 

 longitudinal division, with tl;e halves beginning to separate where attached to the spindle 

 fibers. (After Kiichi Miyake.) 



mother cells, as we found to be the case with the spores of Os- 

 munda, and we must therefore class fern spores and pollen 

 grains as homologues. 



According to the theory of the segregation of homologous pa- 

 ternal and maternal chromosomes, two of the four microspores 

 would possess paternal and two maternal characters from any 

 pair in the grandmother cell. We will defer further discussion 

 of this until the origin of the megaspores has been traced. 



We must now note that competent observers' have found in 

 many instances that in the prophases of the dividing grand- 

 mother cells of microspores and megaspores the nuclear thread 

 does not take the position of two parallel strands, and segment 

 into various side-by-side pairs of paternal and maternal chromo- 



