Latent Characters 231 



their types, never producing true florets. No 

 mark of the original difference was to be seen 

 between them. But last summer (1903) both 

 reverted to their prototypes, bearing rela- 

 tively large numbers of ordinary double flower- 

 heads among the great mass of green stalks. 

 Some intermediate forms also occurred consist- 

 ing of green-scaled stalks ending in small heads 

 with colored florets. 



Thus far we have an ordinary case of rever- 

 sion. But the important side of the phenome- 

 non was, that each plant exactly ^* recollected " 

 from which parent it had sprung. All of those 

 in my garden reverted to the carmine florets 

 with white tips, and all of those in the nursery 

 to the pale orange color and the other character- 

 istics of the ^' Surprise " variety. 



It seems absolutely evident, that no sunple 

 loss can account for this difference. Something 

 of the character of the parent-varieties must 

 have remained in the plant. And whatever 

 conception we may formulate of these vestigial 

 characters it is clear that the simplest and most 

 obvious idea is their preservation in a dormant 

 or latent state. Assuming that the distinguish- 

 ing marks have only become inactive by vires- 

 cence, it is manifest that on returning each will 

 show its own peculiarities, as recorded above. 



Our second point was the incomplete loss of 



