L] ORIGIN OF HERITABLE CHARACTERS 287 



heritable, the quahties which they express having been in some way stamped 

 upon the gametes, so that they are transmitted to the offspring. The hne of 

 distinction between these categories has often been sharply drawn: but in 

 Plants there is reason to doubt whether the difference between them is any- 

 thing more than an expression of limited experience: in fact, whether varia- 

 tions in the first instance apparently of the fluctuating type may not turn out 

 to be in the long run heritable. A Class of Plants with a ver)- long and 

 consecutive history, such as the Ferns, gives an exceptional opportunity for 

 testing this point. 



The question of the origin of heritable characters is still quite an open 

 one for Plants. The fact that in Animals the germ-cells are segregated early 

 from the somatic cells does not affect the question, for in Plants such early 

 segregation does not occur. In them, that is in Plants, the tissues, still un- 

 differentiated as vegetative or propagative, are for long exposed to whatever 

 the conditions of the individual life may have been before its gametes are 

 specialised. This suggests that Plants would be particularly favourable 

 subjects for critical observation. The prevalence in them of parallel de- 

 velopment, or even of convergence in respect of characters that appear to 

 be adaptive, suggests that the changes upon which they have been built were 

 not produced at random. In particular, Ferns with their striking instances 

 of homoplasy offer favourable opportunity for forming an opinion whether 

 or not heritable changes of an adaptive character may have been promoted 

 or actually determined in their direction or quality in some way by the con- 

 ditions, external or internal. Moreover, in Ferns, with their long fossil history, 

 these need not have acted within the restricted time-limits of present experi- 

 ment. The wide latitude of geological time has been available for evolution 

 to proceed. Though we may grant that the direct and immediate effect of 

 external or internal conditions in producing adjustment of structure is not 

 immediately and visibly heritable in the individual, nevertheless it is possible 

 that the effect of such influences continued through long ages may become 

 apparent in the race. It will be well, however, to realise that such a relation 

 of inherited characters to the directive influence of conditions exercised over 

 long periods of time is not recognised in certain quarters, in particular by 

 the followers of Weismann. 



On the other hand, the opinion of de Vries, writing more specially in 

 relation to plants, ma}- be translated thus {^Mutationstheoric, Vol. I, p. 144) : 

 "And so I come to the conclusion that the Mutation Theory demands a 

 mutability of organisms in all directions. Neither Palaeontological nor 

 Systematic facts are irreconcilable with this view: and the construction of 

 ordinary or collective species from groups of elementary species, whose 

 characters diverge from one another in every direction, indicates clearly an 



