92 PLANT MORPHOLOGY 



Replacement and transformation behave as two fluids which are, 

 and two fluids which are not, miscible; in the first case the inner 

 structure is different, and in the second there is a correspondence. 

 The comparison is a limping one, but still gives us a fair illustra- 

 tion. 



As a matter of fact, we do find every intermediate step between 

 stamens and flower leaves, and we cannot doubt that these have 

 come into existence because a stamen, or, in other words, a stamen 

 fundament, has at different stages of its development received a 

 stimulus which has caused it to develop into a flower leaf. We find 

 correspondingly, that the earlier developmental stages of a stamen 

 and a flower leaf are parallel throughout, while in the above-cited 

 example of the branch and a leaf-lobe of a Jungermanniaceous 

 liverwort their developmental histories are throughout different, as is 

 shown by the arrangement of cells. In the case of stamens, therefore, 

 there occurs not a replacement, but a transformation. And, indeed, 

 a limited one. Not any "lateral members" you please may arise 

 instead of stamens, but only and always those which we subsume 

 under the concept leaf, because they evidently have peculiarities in 

 common. Besides, there are also normal flowers which exhibit all 

 intergradations between flower leaves and stamens. The former 

 Coulter and Chamberlain would regard as leaves, the latter not; 

 where, however, is the line of separation between them? 



From the limited power of transformation possessed by organs 

 it results that in causal morphology the problem is, then, not a phy- 

 logenetic, but an ontogenetic one. Whether sporophylls or foliage 

 leaves are the older phylogenetically may be disregarded. For it 

 appears more important first to determine why the power of trans- 

 formation is limited, why a shoot-thorn or a shoot-tendril may be 

 transformed only into a shoot, a stamen or a carpel only into a 

 "leaf;" and second, what conditions are determinative thereto. 



The first step toward the solution of the problem is that we learn 

 to call out experimentally and at will such transformations as we 

 have heretofore occasionally observed as "abnormalities." 



This has been successful in experimental morphology in a great 

 number of cases, and in the future will be still more so. To be sure, 

 we are still unable to induce the transformation of stamens into 

 flower leaves at will, we only deceive ourselves when we believe 

 that the art of the plant-breeder has succeeded in doing this, for in 

 reality all he has done is to isolate such races which have occurred 

 in nature with more or less doubled flowers, and in this regard we 

 stand in contrast to the fungi and insects, the activities of which, 

 as Peyritsch and others have shown, often unconsciously of 

 course call forth such transformations. 1 Yet it has been possible 

 1 The literature is presented in Goebel, Organographie , part i, p. 165. 



