i^oS] YAMANOUCHI—SPOROGENESIS IN NEPHRODIUM 23 



in connection with the blcpharoplast in pteridophytes will be discussed 

 in the next paper, which will deal with gamctogenesis and fertiliza- 

 tion in Nephrodium. 



Uniformity in the number of chromosomes, — Str.\s burger's 



paper (46) entitled '^The periodic reduction of the number of the 

 chromosomes in the life-history of living organisms," which was the 

 first enunciation of the significance of sporogenesis and the reduction 

 phenomena in relation to the altemation of generations in plants, 

 was based upon the discover}^ that there is a uniformity in the number 

 of chromosomes in certain phases of the life-histor}' of the individual. 

 Evidence seems to be accumulating from different groups of plants 

 confirming the notion of constancy in the number of chromosomes in 

 individuals under normal development, although -there are instances 

 where different authors report different counting. 



Generally speaking, accuracy in counting the chromosomes 

 depends upon the number and form of the chromosomes in the 

 individual under consideration. AMien the number of chromo- 

 somes is large and their form is slender and curved, accurate counting 

 is naturally difficult. In Xephrodium the ^Titer found this difficulty, 

 due to the great number and curved form of the chromosomes, but it 

 was not impossible to make reliable counting by using certain stages 

 in which the chromosomes assumed a straight rod shape and a regular 

 arrangement, characteristic of this species. In sporogenesis there 

 are three conditions favorable for counting. First the prophase after 

 the segmentation of the spirem into chromosomes, up to the stage just 

 before the disappearance of the nuclear membrane, was favorable 

 for counting It might well be expected that two parts of the same 

 chromosome which happened to be cut might appear in two successive 

 sections of the same nucleus, and this would make the counted number 

 larger than the actual. Nevertheless, after carefully e^mining the 

 mitosis in this stage, it was possible to find out the secti<Hi5 with the 

 least number of cut chromosomes, sometimes only two, as was given 

 in frgs. 24a^ h^ c and 25a, &, cr, d. The second condition good for accu- 

 rate counting is the polar view in the early metaphase ; and the thir! 

 case is the polar view at a certain stage of the late anaphase, in which 

 the chromosomes are arranged near the pole. 



In the tvpical mitosis of the vegetative ceU, <m the other hand, the 



