THE MALE FERN DRTOPTERIS FILIX-MAS 



One would also expect numerous trivalents in the triploid hybrid, which, however, 

 also do not occur. 



Meiosis in the triploid hybrid between D. Filix-mas and D. abbreviata is shown in 

 Fig. 34^, in which some cells at the first meiotic division are put between comparable 

 views of the two parent species. The irregularity produced by lagging unpaired 

 chromosomes is very conspicuous in the hybrid. The details of chromosome pairing are 

 better displayed in a 'squash' preparation, and comparable cells of the three plants 

 are given again in this technique in Figs. 35-37. Fig. 37 is reproduced at a higher 

 magnification than Figs. 35 and 36 to facihtate observation, and a diagram of the 



CD o ^ 



^ .<% 











o 



¥ 



V>^V .^0 





^ 





^ ^ ^^oo 



o 



Triploid hybrid 3n -- 123 ^ ^U ^ 



Fig. 38. Explanatory diagram to Fig. 37. x 2000. Pairs in black, univalents in outline. 



complete analysis of it is given in Fig. 38. The 123 chromosomes contained in it 

 are represented by 40 pairs, 40 univalents and i trivalent. This is not the type of 

 pairing found in autotriploid Osmunda but is closely comparable to that of allotriploid 

 watercress, allowance being made for the different monoploid number {n = ^i in 

 Dryopteris). The single trivalent is most easily explained by postulating one segmental 

 interchange between two otherwise non-homologous chromosomes in the D. Filix-mas 

 nucleus. Except for this the pairing bears a very close numerical relation to the basic 

 haploid number, and the interpretation would appear to be, as in the case of the 

 watercress, that all the chromosomes of the diploid species can find partners in the 

 tetraploid species but that these represent only half the nucleus of the latter, the other 

 half apparently appertaining to some different species with different homologies in its 

 chromosomes. 



53 



