ALTERNATION OF GENERATIONS 449 



diploid complement of the sporophyte. And, as in regeneration, 

 it produces functional gametes which should yield tetraploid 

 sporophytes in the next generation. This process is recurrent 

 although indefinite repetition is scarcely conceivable. 



When crossed with another race having the same chromosome 

 number {2n = 60) the first generation was normal and its 

 gametophytes all had 30 chromosomes. Yet these produced 

 sporophytes all having about 45 chromosomes. By selfing the 

 gametophytes from these plants and crossing them with normal 

 types it was proved that the male gametes had often undergone a 

 second reduction, to 15 chromosomes {cf. Manton, I.e.). One- 

 quarter of the second generation were aposporous plants, and all 

 these bred true to the character of having a redundant although 

 apparently often imperfect meiosis. The combination of apospory 

 with antheridial meiosis gave chromosome numbers increasing by 

 about a half in succeeding generations, thus : F^, ca. 45 ; F^, ca. 65 ; 

 F4, ca. 95. These results point to the redundant meiosis being an 

 independent hereditary character. It is analogous to the condition 

 in the pseudo-hermaphrodite Icerya purchasi. The reduction to 15 

 indicates that the original plants were tetraploid. 



In certain "normal" derivatives of the mutant the spores have 

 been found to be replaced in part by spermatozoids (Gairdner, 1933). 

 The traditional homologies which apply throughout the higher plants 

 are thus overthrown at one step. Such revolutionary behaviour can- 

 not perpetuate itself, since no corresponding archegonia are formed. 

 The stabiHty of the alternation of generations in the higher plants 

 and the gradualness of its transformation is thus seen to depend, 

 not on the absolute and mystical distinction which morphologists 

 have made between the successive generations, but on the impossi- 

 bility of making a decisive change in any one stage of development 

 without making an exactly adapted change in others at the same 

 time ; in other words without a coincidence in mutations such as 

 does not occur in nature. The stability of the distinction between 

 alternating generations depends on the complexity of their mutual 

 adaptation and not on a profound genetic difference between them. 



The unbalanced occurrence of apospory and parthenogenesis, as 

 well as many observations on polyploids, show that the association 



R.A. CYTOLOGY. 15 



