344 Varietäten, Descendenz, Hybriden. 



O. Lamarckiana the mutability for dwarfs, therefore, must have 

 increased at least tenfold, and for gigas types about sixfold. The 

 material cause for this improvement is in all probability the same 

 as or closely connected with the cause of the largely increased 

 number of mutative forms which are known to Start from O. La- 

 marckiana, 



From the cross 0. biennis mut. nanella X O. ö/ewms only dwarfs 

 of a uniform type arose (108 Ex.). O. biennis X O. biennis mut. 

 nanella was in the first generation exactly like pure biennis; O. 

 Lamarckiana X O. biennis mut. nanella exactly like O. Lamarckiana 

 X O. bieyinis. 



O biennis semigigas is self-sterile, but when poUinated by 

 O. biennis gives for one-half pure biennis with 14 chromosomes, and 

 for the other half a new, slender type with 15 chromosomes. 



O biennis mut. sulfiirea easily yields constant races ofa uniform 

 sulphur color. 



The question whether there is any causal relation between 

 partial sterility of the sexual cells, hybridism, and mutability has 

 to be studied in all those instances in which mutations are known 

 to occur or to have occurred. In some of these cases, at least, the 

 conditions are far more simple than for the evening primroses, as 

 for example in Capsella Bursa-pastoris. 



The mutative condition of O. biennis may be ascribed to some 

 "germinal disturbance" of its hereditary qualities. Or, if we replace 

 this vague and meaningless expression by a Sharp hypothesis, we 

 may assume as its cause the presence of one or more pangens in 

 a labile position. The transition from biennis to Lamarckiana would 

 then require the addition of one or more pangens in the same 

 State, in order to explain the higher percentage of mutants and 

 the larger number of their different forms. The presence of such 

 labile pangens seems well proven by theresults of numerous crosses. 



The contention, however, that the translation of "undisturbed 

 germinal material into a State of disturbance" or of one or more 

 pangens from the stabile into the labile condition, may be induced 

 by external influences in pure species, has not as yet found general 

 acceptance. Some authors believe that crosses between different 

 types are required to secure this effect. At this moment, it seems 

 difficult to give experimental evidence for or against this view. Until 

 this is reached, we must rely upon comparative studies in order 

 to answer the main question whether or not the observed mutations 

 in the evening primroses are analogous to those by which the 

 mutation theory explains the evolution of the animal and vegetable 

 kingdoms. 



The mutants of 0. Lamarckiana all agree with that species in 

 certain characters, and not one of them shows any indication of 

 a reversion toward any of the allied wild types. If the mutability 

 was an effect of crossing, some marks, at least, of the other parent 

 would be expected to reappear. 



Besides this consideration, the available evidence lies in the 

 fact that the derivatives of O. Lamarckiana^ originated in the 

 author's garden, differ from one another in marks, which are, 

 although not identical, strictly analogous to those which differentiate 

 the wild species of the whole group. In some cases the differences 

 are even larger. Those between the wild species are often very 

 small and limited to certain life periods, leaving the species quite 

 alike during the remainder of their development. No arguments 



