8 • Background 



tion until the effects on flowering can be ascertained. The situations 

 may be complicated by the fact that conditions bringing about 

 initiation are not always the same as those favoring bud develop- 

 ment, and these in turn may differ from those required for anthesis. 

 As mentioned earlier, experimentalists have been most concerned 

 with initiation; since, however, flower primordia in their earliest 

 stages are detectable only by dissection and microscopy, the data 

 in many studies have been based on the appearance of macroscopic 

 buds or flowers. 



Within this general framework, methods of evaluating the 

 results quantitatively are less obvious and vary considerably. The 

 crudest method is simply to record the time required for the first 

 appearance of the designated floral stage in the various treatments. 

 This of course will vary between individuals given the same treat- 

 ment, so averages are used. Alternative but related data are the 

 percentage of plants in each treatment showing the designated 

 stage at a given time after the start of the experiment. There are 

 also plants, such as soybean (Glycine), in which flowering may 

 occur at a number of nodes, and the effectiveness of treatments can 

 be estimated by establishing the average number of nodes with 

 flowers per plant after a given time. Still another related method 

 is that of assigning arbitrary number values to various stages in 

 the development of flower or inflorescence primordia. With a scale 

 so established and an appropriate time for evaluation chosen, the 

 plants in each treatment are dissected or examined and the result- 

 ing values averaged. A danger of this method lies in the subjective 

 judgments involved in assessing stages and assigning values to 

 them. 



These procedures are all related in that the major independent 

 variable, other than the nature of the treatments given, is time. 

 That is, in a graph of results so obtained, each flowering value, 

 however stated, is a function of time in or after treatment. A draw- 

 back of such methods is that if the treatments differ in their effects 

 on overall growth, and the times involved are (as is usual) a week 

 or longer, differences in flowering values may simply reflect differ- 

 ences in growth rate of the entire plant. For example, a 10° C 

 increase in temperature might double the rate of vegetative growth 

 and also that of the appearance of buds. But in such a case the 

 rate of bud appearance relative to vegetative growth is unchanged, 

 although time-based data would indicate more rapid flowering. 



