Growth and variation in maize. 139 



while it is true that ou the average, the relatively small plants tend 

 to become larger as the season advances, it is equally true, as brought 

 out in the preceding section (cf. p. 120), that such small plants tend to 

 remain smaller than the average for the season as a whole. Veiy 

 similar statements may be made for plants which were relatively large at 

 the beginning of the season. 



In several of these figures there is still another condition worth 

 noting. Thus, as is most clearly seen in Fig. 12, there is a period 

 during the middle of the season in which the mean quintUe position of 

 plants starting in quintile I and V remain practically constant. In 

 these cases there is a sharp change in the mean quintile position at 

 the beginning of the season and a similar change at the end of the 

 season. A similar, although less marked, condition is seen in Fig. 11, 

 while in Fig. 10 the conditions are quite different. The data are not 

 sufficient to discuss the possible meaning of these variations in the series. 



In tables 31 to 46 there is given in addition to the mean quintile 

 position the standard deviation of this mean position. It may be seen 

 from these constants that in general there is an increase in this standard 

 deviation with the advance of the season. The fluctuation in these 

 constants due to the small number of individuals in each table are so 

 great that it is not possible to draw further conclusions from them 

 with certainty. The constants have been included in the tables because 

 they give some idea of the probable error to which the mean are subject. 



b) Plants Ending in a Given Quintile. 



In tables 46 — 60 there are given the mean quintile positions and 

 standard deviations for plants ending in the given quintiles. These 

 may be examined very briefly and by the same methods employed in 

 the preceding paragraphs. Fig. 13, 14 and 1.5 show the graphs of the 

 means for the several groups of plants in each series. 



Comparing these figures with Figs. 10 to 12 it is seen that in 

 general there is a much greater tendency here for the mean quintile 

 position of each gi'oup of plants to lie near the mean of the popu- 

 lation, ^iz., 3'0. This is particularly marked in Fig. 15. Here, with 

 the possible exception of plants ending in quintile I, these values 

 fluctuate about this mean value with only slightly greater deviations 

 than might be expected on the theory of chance alone. In Figs. 13 

 and 14 the fluctuations are somewhat farther from the mean but in 

 general they are not so far as in the corresponding Figs. 10 and 11. 



