236 KIYOSHI masüi. 



Hance ('17 a) also made a similar olservatiou iu the mammalian tissues. 

 He snys : " Synizesis was uüt seen in this insxterial excapt iu tlie center of a 

 piece of tissue -which was rather larger and where the fisati\e had not 

 jxinetiated." Thus he attributes the cause of th.9 polarization of the spiremes 

 to the result of imjwrfect fixation of tissues. Again he says : " Even here it 

 is not the tight ball of threads figured so many but gives every evidence 

 that it is the result of the extraction of the fluid. In many preparations 

 where any evidence of the synizesis of tliiu chromatin thi-aads occurs, there is 

 only a slight contraction if the threads away from the nuclear wall appearing 

 as though the fluid which had suj^ported these thi-eacls had been removed. 

 The slffinkage appears to hä equal from all sides, although occasionally a 

 cell is foxmd with the chro7natin threads massed at one side. In well teased 

 or siuiül piece of tissue these same stages apjÄar with the tlu-eads well 

 separated and there is no shrinkage away from the nuclear wall." 



But iu my materials the oljservation of the preparations comjielled us to 

 suhiiii the following facts: 1. Whatever fixations may le used, the synizesis 

 (conti'action figure) can Ije seen in every part of the sections, though it is very 

 slight in the mouse. 2. This phenomenon occurs m the definite jxiritxl of 

 the growth stage of the spermatocyte, and in very young stage, but never 

 in other stages. 3. At the end of this st:xge, the chromatin spiremes appear 

 in about half the original numljer and twice as thick as the leptfjtene stage. 



Judging from these facts it seems most probable that the -pohi aggrega- 

 tion of the chromatin thi'eads in the rabbit and iu the mouse represents the 

 normal state which occurs in the definite period of the s^iermatocyte and 

 coiTesponds to that of the other mammals. Moreover it is a sticking evidence 

 iu support of this view, tLit in the rabbit, throughout the wall of the tuliule 

 in which the cells are presenting the synajBis, the parallel aiTangements of 

 the clu'oraatin tlireads, without exception, show a dc<3ided canti"action t>f tlie 

 thrciwls. 



As I have stsited of the horse iu the previous jmper ('19), the leptotane 

 tlireads iu tlie rabbit arc aggrcgat*.>d to one pole of the nucleus where the 

 idiozome is situated. The same phenomenon was described by 'WoDSED.viiEK 

 ('13) in the pig and Jordan ('12) in oix)SSum. Morse ('09) foimd a simiLir 

 cionditi>u of the condensation of the chromatin threads in the sjx^rmatücyto 



