206 



OF THE CHROMOSOMES 



' S I 



duced number appears in all the cells of the sexual generation (pro- 

 thallium, pollen-tube, embryo-sac), beginning with the mother-cell of 

 the asexual spores from which this generation arises. 



In every case we must distinguish carefully between the primary 

 pseudo-reduction in the number of chromatin-masses, and the actual 

 reduction in the number of chromosomes ; for the former is in some 

 cases certainly not an actual halving of the number of chromosomes, 

 since each of the primary chromatin-rods is proved by its later history 

 to be bivalent, representing two chromosomes united end to end (sal- 

 amander, copepods). In these cases the actual reduction takes place 



in the course of the last two 



A 



divisions (formation of the polar 

 bodies and of the spermatids), 

 each bivalent chromatin-rod di- 

 viding transversely into the two 

 chromosomes which it repre- 

 sents, and at the same time 

 (or earlier) splitting lengthwise. 

 Each primary rod thus gives 

 rise to a tetrad consisting of 

 two pairs of chromosomes which, 

 by the two final divisions, are 

 distributed one to each of the 

 four resulting cells. In the 

 copepods the first division sepa- 



A. Ascaris-type. Double longitudinal split- rates the longitudinal halves of 



ting of the primary rod; no reduction in the the chr omoSOmeS and IS there- 

 number of granules (ids ). B. Copepod-type. 



A longitudinal followed by a transverse division fore an " equal division " (WeiS- 



f , th ! K pri T ary rod ' the . number of s ranules mann). The second division 



halved by the second division. 



corresponds with the transverse 



division of the primary rod, and therefore is the " reducing division " 

 postulated by Weismann. 



This result gives a perfectly clear conception of the process of 

 actual reduction and its relation to the preparatory pseudo-reduction 

 that precedes it. It has, however, been absolutely demonstrated in 

 only two groups of animals, viz. the copepods and the vertebrates 

 (amphibia), and a diametrically opposite result has been reached in 

 the case of Ascaris (Boveri, Hertwig, Brauer) and in the plants (Gui- 

 gnard,-Strasburger). In Ascaris typical tetrads are formed, but all 

 observers agree that they arise by a double longitudinal splitting of 

 the original chromatin-rod. In the plants no tetrads have been ob- 

 served, but the precise nature of the maturation-divisions is still in 

 doubt. 



We have thus two diametrically opposing results. In the one 



Fig. 102. Diagram contrasting the 

 modes of tetrad-formation. 



