General Notices. 316 



reserve the name of atavism ; then we must consider separately, and in a 

 more special manner, the attraction or form of resemblance to the actual 

 parent, which, less powerful, but acting at a smaller distance, tends to per- 

 petuate in the progeny the characters peculiar to the immediate parent. So 

 long as the parent has not departed in a material degree from the type of 

 the species, the two forces act parallelly, and are blended ; and the variations 

 whicli may follow in such a case, through the law of idiosyncrasy, may pre- 

 sent themselves indefinitely in all directions, without exhibiting a peculiar 

 leaning to any one. It is difl^erent when the actual parent has varied in a 

 marked manner from the type ; the force of resemblance to tlie actual parent 

 then becomes combined with that of the individual variations, and the result 

 is an excess of deviation in the direction of the resultant of these two com- 

 bined forces ; or, if it be preferred, the individual variations will then 

 radiate, not round the common centre of the type, but round a point situated 

 on the line which separates the type from the variety first obtained. Aban- 

 doned to nature, individual variations almost always perish in the super- 

 abundant mass of individuals which are unceasingly undergoing destruction. 

 Hence the fixity of natural species. But when collected by man, these 

 variations are protected, their descendants become multiplied ; obeying then 

 the more complex laws Mhich govern them, they produce their numerous 

 modifications, which man has been able to fix for his own service. It is 

 thus also that the influence of man, in exclusively choosing modified indi- 

 viduals, to multiply the progeny, supplies a counterbalance in his constant 

 efforts to tlie equally constant force of atavism, and thus succeeds in eman- 

 cipating or fixing modified races. From the preceding considerations it will 

 be evident that one of the points, which we regard as most essential, con- 

 sists in combating, as efficiently as possible, the force which we have called 

 atavism. Now this force, less direct in a manner than that of resemblance 

 to the immediate parent, may perhaps act with great persistence. If I might 

 use another comparison, borrowed from the laws of mechanics, I would say 

 tliat on account of the distance of its point of origin, its decrease takes 

 place in an almost imperceptible measure, in the course of the small number 

 of generations on which man can exercise influence ; while the decrease of 

 the other force (that of resemblance to the actual parent) goes on in geo- 

 metrical progression. I have therefore been led to set up a theory which, 

 however, I only present liere now with extreme hesitation, on the subject of 

 the course to be pursued in endeavoring to obtain varieties from a plant not 

 yet subjected to modification. To obtain from a plant not yet modified 

 varieties of a kind determined beforehand, I first apply myself to producing 

 variation in any direction, selecting for the reproducer not that one of the 

 accidental varieties which approaches nearest to the form I wish to obtain, 

 but simply that Avliich differs most from the type. In the second generation 

 the same care would lead me to choose a deviation the greatest possible at 

 first, then the most diff^erent from tliat which I should have chosen in the 

 first place. The necessary result of following tliis course through several 

 generations is an extreme tendency to vary in the product thus obtained. A 

 further result — and this is the principal point in my opinion — is, that the 



