104 FREDERICK ORPEN BOWER 



consideration this necessary fact of descent, that propagative 

 organs must have appeared in each completed cycle. Propaga- 

 tive organs are not things formed de novo in descent by transforma- 

 tion of some preexistent vegetative structure — they are organs sui 

 generis. It was Goebel who first realised this, and his statement 

 of the principle for sporangia in 1880 may be held to have consti- 

 tuted an essential step forward in modern morphology. For us 

 at present the chief interest lies in this: that the theory of meta- 

 morphosis left out of account this essential feature of descent, 

 that the propagative organs are not afterthoughts, but recur as 

 an incident in every completed life cycle. 



The foregoing paragraphs are not written in any frame of cen- 

 sorious superiority. The object has been to learn from experience 

 of the past the methods to be used in the present and the future 

 in the quest for phyletic relations, and we are now in a position 

 to recognise some at least of the sources of weakness or of strength 

 in any phyletic view that may be promulgated. It is the absence 

 of direct evidence, due to the imperfection of the fossil record 

 on the one hand, and to the survival of so few representatives of 

 earlier evolutionary steps on the other, that has made it thus 

 necessary to fall back upon elaborate weighing of evidence, and 

 has left the conclusions so much matters of opinion rather than of 

 actual demonstration. But clearly seeing that this is so, we shall 

 now recognise the following sources of weakness under which views 

 as to descent are liable to suffer. 



1. Assumption. Few realise as they write how largely assump- 

 tion figures in their arguments. A careful analysis of the lines of 

 reasoning is often necessary to detect where an assumption lies. 

 Unfortunately such analysis has often to be made by a critic 

 instead of by the author himself. We have seen an instance in 

 the assumption that all that is simple is primitive, and the con- 

 sequent conclusion that the Leptosporangiate state preceded the 

 Eusporangiate. It is well to bear constantly in mind that the 

 simplicity of reduction is very prevalent. This is much more 

 generally recognized now than formerly. There is even a con- 

 verse danger, of which modern instances could be quoted, of 

 assuming reduction in order to explain facts that are otherwise 



