INTRODUCTION 15 



must therefore look to other sources of information in 

 our quest for the ancestors of the present flora of the 

 earth. 



Much can be learned as to the relationships of plants 

 from a study of their external structure, and the classi- 

 fication, especially of the higher plants, is based largely 

 upon purely external characters. While such charac- 

 ters are usually reliable when dealing with nearly re- 

 lated forms, they are likely to be misleading when we 

 try to trace out the affinities of plants whose kinship 

 is not so obvious. Here it is important to take into 

 account, for comparison, the more obscure points of 

 structure, - - for it not infrequently happens that re- 

 semblances may thus be traced which are not evident 

 at first sight. Thus, in comparing the Mosses and 

 Ferns, it is the minute reproductive organs and em- 

 bryos which show the unmistakable relationship of 

 these plants, while their more conspicuous external 

 structures are very different. 



There is little question that, as in the study of animal 

 forms, it is the careful investigation of the life-history 

 of the plant which affords the surest clue to its affinities 

 with other forms. The generally accepted view that 

 in animals the developing germ repeats in a general 

 way the evolution of the race, is also applicable, in 

 some degree at least, to plants, and by far the most 

 important discoveries, with reference to the origin of 

 Dlant forms, have been due to studies of this nature. 

 Very often the early stages of the embryo and repro- 

 ductive organs in different plants reveal resemblances, 

 while the adult stages may have, apparently, very little 

 in common. These embryonic phases are less affected 



