THE LOW TEMPERATURE PROMOTION OF FLOWERING 61 



kinds of substances showing gibberellin activity change in various 

 species during the induction period. 



For many years after the hormone concept of flowering was 

 developed, it was felt that proof of the theory would be the demon- 

 stration that flowering plants could be extracted, and the extracts 

 applied back to vegetative plants causing them to flower. Extraction 

 of vegetative plants should yield inactive (or weaker) extracts. This 

 has now been clearly achieved with the gibberellins. Are the gib- 

 berellins, then, identical with the flowering hormone? Obviously not 

 for short-day plants. It was suggested that there might be two hor- 

 mones, one lacking in vegetative cold-requiring or long-day plants 

 (gibberellins) and one lacking in vegetative short-day plants. Since 

 a flowering short-day plant will cause a vegetative long-day graft 

 partner to flower, both hormones would have to be required for 

 flowering; but if that is the case then a vegetative short-day plant on 

 long days, having the gibberellin but lacking the short-day product, 

 should cause a vegetative long-day graft partner to flower on short 

 days. This seldom occurs, but the above grafting experiment of 

 Melchers is nearly an example. 



Could the vernalin of Melchers be gibberellin ? This does seem 

 possible, although there are also a number of difficulties here. Treat- 

 ment with gibberellin leads first to active cell division in the subapical 

 region of the rosette axis, then to the formation of an elongate axis, 

 and only secondarily to flower formation. Flowering in response to 

 cold or to long days, however, occurs before or concurrent with stem 

 elongation. Perhaps the activities associated with elongation of the 

 stem lead to flower formation, and gibberellins only cause the stem 

 elongation. Even this theory is not completely satisfying, but it does 

 seem to be the most likely explanation presently available. Production 

 of gibberellins in response to cold- or long-day could at least be part 

 of the flowering process ; a part capable of making the entire process 

 begin if it becomes predominant. 



Very recently, Jan A. D. Zeevaart and Lang (81) at Pasadena have 

 performed an elegant series of experiments which demonstrate that 

 in Bryophyllum, the long-short-day plant mentioned above, applica- 

 tion of gibberellin leads to the formation of flowering hormone on 

 short days but is not the hormone itself. Plants which had been 

 induced to flower by application of gibberellin on short days would 

 induce vegetative graft partners to flower even on long days. Since 



