248 REPRODUCTION, EVOLUTION, ETC. 



became interpolated by a gradual delay in the occurrence of meiosis. 

 Against this possibility it may be pointed out that 



(i) There are very few Phaeophyceae in which the haploid 

 generation is wholly dominant. It is possible, of course, that they 

 were more numerous and have subsequently been displaced by 

 the more recent types in which the diploid generation plays a more 

 significant role. 



(ii) Ectocarpus siliculosus in its English and Mediterranean forms 

 would both begin and end the series, and this hardly seems con- 

 ceivable. This, however, could not be regarded as a fundamental 

 objection because it might equally well be argued that the species 

 forms an excellent example of how the process of interpolating the 

 diploid generation took place. 



(iii) The frequency of parthenogenesis in the Ectocarpales 

 suggests decadence of sexuality rather than the existence of a 

 primitive condition, but it could also be argued that there is a 

 decadence of sexuality in the Laminariales and Fucales. 



B. Plants with only a diploid generation, e.g. Fucales, are the 

 most primitive, and the haploid generation has been interpolated 

 subsequently. If this interpretation is correct the only obvious 

 source of origin for the group would be from the Siphonales be- 

 cause a flagellate ancestry would be most unlikely under such 

 circumstances. The evidence that might be adduced in support of 

 this hypothesis is tabulated below : 



(i) All the Fucaceae are diploid, and these form a large pro- 

 portion of the Phaeophyceae and also have an extremely wide 

 distribution. 



(ii) In the Laminariales the diploid phase is dominant. 



(iii) The haploid phase is frequently omitted in Dictyota (cf. 

 p. 165) and also in Cutleria. 



(iv) The majority of the macroscopic filamentous forms are 

 diploid, the small ectocarpoid filaments forming the haploid 

 generation. 



One important objection to this view is the concomitant require- 

 ment that the early Phaeophyceae must have started life with a 

 highly complex structure, e.g. Fucus, though of course some such 

 structure can be found in the Siphonales. It must also be re- 

 membered that the interpolation of a diploid generation into an 



