326 AUGUST WEI SM ANN [april 



occasionally over-estimated the bearing of a newly acquired item of 

 knowledge in the joy at its discovery, — that is, I have looked at it 

 from too one-sided a point of view. This was the case in my paper 

 on the meaning of sexual reproduction, in which I represented this 

 factor in individual variation as its real and main root, while it is, I 

 now believe, only the indispensable means of mingling these variations 

 and of their persistent renewal. It is immaterial whether I myself or 

 one of my sharp-sighted critics was the first to discover this mistake ; 

 but in any case it gave me occasion to sink a new ladder into the 

 mine of investigation, to bore for the deeper ore-stratum of the source 

 of variation, and to find it in the minute local fluctuations in nutrition 

 which strive to disturb the equilibrium of the system of determinants 

 in the germ-plasm. In this way I arrived at the theory of " germinal 

 selection," which, although recognised only by a few — publicly, I 

 believe, only by Emery — as a justifiable hypothesis, will, I should like 

 to believe, yield greater results in the future. 



It has also been said that I have carried my theories much too 

 far, and have lost myself in details and speculations. Ironical 

 admiration has been bestowed on the courage which enabled me to 

 apply my theories to particular problems, and which did not shrink 

 from the many auxiliary hypotheses necessitated by applying the 

 main theories to special questions. To speak thus is to overlook the 

 fact that nothing but persistent pushing of a guiding hypothesis to its 

 furthest limits leads to a recognition of its defects and weaknesses, and 

 so paves the way for farther progress. It is quite easy to set up 

 theoretical principles, to propound a " theory of heredity " or of 

 ontogeny, if one confines oneself to elucidating only the most general 

 phenomena. The test of the usefulness of theories lies in applying 

 them to special phenomena ; it then becomes apparent whether and 

 where they touch improbabilities, and where new facts are required to 

 improve them or to replace them by others. It is not the joy of 

 '•' explaining everything," but the desire to see them tested on all sides 

 and to give them freely into the hands of my critics, that has led me 

 to apply my principles of explanation to particular problems and to 

 carry them as far as possible. This has been well understood by Yves 

 Delage 1 — by far the most objective of my hostile critics — in the following 

 passage : " Enfin il faut savoir gre a Weismann d'avoir ete jusqu'au 

 bout des consequences logiques de son systeme. 11 a tcnii a tout 

 expliquer, et il n'a pas recule" devant la necessite de compliquer sa 

 conception fondamentale, si simple et cependant deja si feconde, pour 

 rendre compte des faits de bourgeonnement, de regeneration, de 

 polymorphisme, etc. II aurait, en evitant d'en parler, comme tant 

 d'autres, echappe' a de graves objections ; il a prefere les subir que de 

 reculer devant les difficultes." 



If it be asked why I did not keep silence altogether if many of 



1 "La structure du proto}>lasma et les theories sur l'heredittV' etc., Paris, 1895, p. 708. 



