On the Inheritance of Acquired Characters. 141 



fluence on the monstrosity whether we have regard to 

 the carpels themselves or to the seed which they contain. 

 In the experiments in question their operation begins 

 with the germination of the seeds which are to produce 

 the plants for the experiment. The degree of develop- 

 ment of the monstrosity in such cases would be called 

 an acquired character in the usual acceptance of the 

 term. There is no obvious reason why we should not 

 apply this term to the improved value of the seeds ob- 

 tained in this experiment — an improvement that has been 

 brought about by the very same external conditions 

 which affected the monstrosity. 



The parallel between individual strength and the de- 

 velopment of the monstrosity is not absolute. On the 

 contrary there is a very simple means of separating the 

 tw^o and of obtaining plants of great vigor but almost 

 destitute of supernumerary carpels. This process rests 

 on an accurate determination of what we may call the 

 impressionable period in the development of the mon- 

 strosity. If we examine the young buds under the micro- 

 scope at intervals, we find at about the sixth week from 

 the germination of the seed the stamens and supernumer- 

 ary carpels represented by small protuberances on the 

 growing summit. The relative number of the two struc- 

 tures is obviously settled at this stage; all attempts to 

 alter their relative proportions by subjecting them to 

 different conditions, after this stage has been reached, 

 have been without avail. The critical stage occurs, then, 

 during the first few weeks of life ; so that it is during this 

 period alone, if ever, that it should be possible to sup- 

 press or at any rate to reduce the monstrosity. I have 

 succeeded in doing this, simply by transplanting the 

 young seedlings at the proper time, that is when they 



