342 SCIENCE PROGRESS. 



Vascular Cryptogams having arisen independently of one 

 another must be considered. Such a view, while it does 

 not prevent the use of the stages of complexity of the 

 Bryophyte Sporogonium as illustrations of the probable 

 steps by which the Pteridophyte Sporophyte was evolved, 

 suggests the alternative course of looking for evidence 

 which may indicate how the latter could have been derived 

 directly from an algal form. 



The modifications which the recognition of this possi- 

 bility would entail in current views on homology cannot 

 be touched upon here farther than to point out that the 

 comparison of those forms of sporophyte which there are 

 good grounds for concluding are homogenetic * with one 

 another must be relied upon in preference to comparisons 

 between forms which may be merely homoplastic. 1 General 

 impressions, gathered from a wide survey of the relation of 

 sexual and asexual generations in the vegetable kingdom, 

 must be checked by comparisons limited as far as possible 

 to closely related forms. 



These considerations suggest further investigation of 

 the behaviour of the sexual generation of the vascular 

 plants in the hope that a more accurate knowledge of the 

 changes which ensue on exposure to altered conditions of 

 life may aid in arriving at conclusions as to how the sporo- 

 phyte might have been evolved from organisms which in 

 form, and possibly in physiological properties, resembled 

 the gametophyte of existing plants. The changes of con- 

 ditions to which most importance should be attached are 

 such as may reasonably be supposed to have occurred 

 during the evolution of land plants. From this point of 

 view, considerable importance may be attached to apogamy, 

 and in a less degree to apospory. 



The existence of a nuclear distinction between the. two 

 generations is not necessarily inconsistent with such an 

 origin of the sporophyte from forms homologous with the 

 sexual generation. For it is an assumption that the nuclear 



1 Lankester. The extension of the use of these terms and of the 

 distinction which they imply would do much toward clearing our ideas on 

 many morphological questions. 



