INDUCTION. 119 



independent analogies; (3) the possibility of 

 deducing consequences from the hypothesis, 

 and verifying them by observation or experi- 

 ment; (4) the possibility of deducing the cause 

 or principle as an effect of another still more 

 general. 



Huxley advised Darwin not to publish his 

 doctrine of Pangenesis ; but he nevertheless did 

 publish it, and gave as his reason the pressing 

 importance of co-ordinating the inexplicable 

 facts of inheritance. Other hypotheses have 

 followed Darwin's with as little success. Dar- 

 win did not formulate his hypothesis to support 

 his other theories, but its character was at least 

 in part determined by the latter. It is interest- 

 ing to note that the latest hypothesis of inherit- 

 ance, the most aggressive that has yet arisen, 

 has been developed as a special support for the 

 belief that natural selection acting upon con- 

 genital variations is the sole cause in the pro- 

 duction of species, and that acquired characters 

 are not inherited. The principal evidence on 

 which the hypothesis relies for support, apart 

 from the refutation of the direct evidence ad- 

 duced to support the belief in the inheritance 

 of acquired characters, are the karyokinetic 

 processes in cell division, and the early stage 

 in development at which, in some animals, the 



