80 THE FLOWERING PROCESS 



they are not truly random, and in a technical sense statistical evalua- 

 tions cannot be applied to the results. We have found such an 

 arrangement essential, however, since it is often necessary to find 

 plants of a given treatment in a short time interval, and one cannot 

 search through a truly random distribution. Furthermore, it would 

 be quite amazing if the environment in the greenhouse varied in such 

 a way that one treatment, with plants lined up diagonally, would 

 have a significantly different environment than others in diagonal 

 lines next to it. The distribution of plants may not be random, but 

 it is quite different from the distribution of environmental variables 

 on the greenhouse bench. 



Statistics have seldom been used in the evaluation of experiments 

 such as those described in the ensuing chapters, although use of 

 simple tools such as the standard error of the mean has often helped 

 to define the variability. As a rule, there is a clear cut difference 

 between controls and treated plants — one may flower at a high rate 

 and the other may not flower at all. As can be seen from many of 

 the figures in this book, the experimental points (the actual data) do 

 not always fall on the curves as drawn, but it is equally true that the 

 trends are quite clear. 



III. Experimental Methods 



1. The Darkroom 



As mentioned earlier, dark curtains may be used on the greenhouse 

 bench. This is a standard procedure for maintaining long-day plants 

 in the vegetative condition, and of course short-day plants may be 

 induced in this way. With the cocklebur, however, a single long 

 dark period is sufficient, and in most cases, it is quite advisable to 

 only use a single dark period,^ in which case plants may be wheeled 

 into a dark room where temperature and humidity can be controlled. 

 In our experiments we usually maintain the temperature around 

 23°C, which is near the optimum for flowering, although temperatures 

 from 20 to 30°C are quite satisfactory. If chemicals in solution are 

 applied to plants, it is advisable to maintain a high humidity, since 

 this seems to help in penetration of the chemical into the leaf. 



5 Of course, if the number of dark periods is the variable being studied, this 

 would not be true, and in amateur studies more than one dark period will increase 

 the flowering rate and often make the results easier to observe. 



