BIOSYSTEMATIC STUDY OF SOME GLYCERIA SPECIES -II 



303 



Most cells had one quadrivalent, and about half had one 6-valent as well, the remaining 

 chromosomes pairing as bivalents. Associations with odd numbers were rare, probably- 

 representing stages in the disjunction of more complex multivalents (Figs. 9, 10). 



Numerical unbalance occurs at first anaphase since in a few cells odd numbers of 

 chromosomes, usually 19 : 21, were seen; 18 : 22 were observed in one cell of panicle No. 

 4. Most 6-valents, and several of the quadrivalents, were non-disjunctional^ in type, which 

 would lead to genetical unbalance in the genomes of the pollen grains. Second division 

 proceeds regularly and tetrads are formed, the contents of the pollen then degenerate 

 rapidly. The plants were also female sterile since no seed was produced. The genetical 

 and numerical unbalance does not seem to be extensive enough for its effects to account 

 for the complete sterility of this clone, which suggests that some further genetic and 

 physiological disturbances are present. 



G. X pedicellata 



Pairing at first metaphase was studied in plants from 4 stands of G. X pedicellata. 

 In all the cells analysed pairing was at its zenith, and the number of bivalents recorded 

 represents the maximum amount of pairing in the cell. The frequency of chromosome 

 pairing is shown in Table 4. 



Table 4. 



G. X pedicellata all fixations 2n = 40 











lean metaphase 



' pairing per cell 





No. of 



No. of 













Herb. No. 



Plants 



Cells 



I 





II 



III 



IV 



410 



10 



30 



16-9 





10-7 



0-83 





429 



6 



30 



15-7 





10-9 



0-91 



0-03 



438 



10 



30 



15-6 





11-2 



0-74 





453 



10 



30 



16-6 





10-3 



1-10 



0-06 



Total 



36 



120 











Average per cell 







16-2 



10-7 



0-89 



0-05 



The frequency of chromosome pairing is low, the number of bivalents ranging from 

 six to fifteen per cell, (Fig. 3) trivalents were infrequent, less than one per cell, and only 

 two quadrivalents were observed in all the cells studied. Anaphase was very irregular, 

 the univalents either being distributed precociously and at random to the poles or lagging 

 and dividing at the equator (Fig. 4), and were lost at second division. Considerable 

 numerical unbalance occurred in the pollen, as a result of first metaphase behaviour, 

 since the number of chromosomes at the poles at the end of first anaphase varied widely 

 from 14 to 25, plus, in some cases, one or more divided univalents. Lagging univalents 

 often persisted at telophase and were lost; consequently the number of chromosomes at 

 each pole cannot often be counted. 



It appears that the complete sterility of this hybrid is accounted for by numerical 

 and genetical unbalance in the pollen mother cells and ovules. 



