AGRICULTURE: REED AND HOLLAND 
143 
nal factors in determining the growth, reproduction and senescence of the 
plant, factors which are so potent that they overbalance external factors so 
long as the latter do not too closely approach minimum or maximum values. 
That the oncoming of reproductive processes induces changes in the growth 
rate of the organism is a fact too well known to require comment, but is well 
illustrated by the behavior of the sunflower. 
It appears that growth, its rate, its grand period, and, to some extent, its 
amount are so steadily controlled by factors inherent in the genetic constitu- 
tion of the sunflower that these factors are prepotent unless the external 
conditions depart widely or repeatedly from the optimum. Plants in this re- 
spect are more sensitive to variations in their external environment than ani- 
mals, yet these studies show that even plants are not entirely dependent upon 
environmental (external) conditions for determining their growth rate. 
It may be of interest to inquire whether the formula of autocatalysis applies, 
as well to the smaller plants as to the medium and large plants of this group 
and how the mean values of K for different groups agree. 
TABLE 3 
Height and Growth Constants of Plants Ending in Their Growth in Different 
QUARTILES 
QUARTILE 
I 
II 
III 
IV 
198 cm. 
238 cm. 
272 cm. 
312 cm. 
0.0440± 0.0011 
0. 042 1± 0.0016 
0.0429=t0.0017 
0.0443=^0.0023 
Standard deviation of 
mean value oi K 
0.0052=^0.0008 
0.0079^0.0011 
0.0079=^0.0012 
9.0111±0.0016 
As a basis of classification we divided the plants into quartiles, based upon 
their heights at maturity. Quartile I, contained the smallest plants, quartile II, 
the next larger and so on. Since 58 is not exactly divisible by 4, the quartiles 
were not exactly of equal size: quartiles I and III contained 15 plants each and 
quartiles II and IV contained 14 each. An average of the heights of each 
group of plants at each time interval, t, gave a corresponding value of x, from 
which the several values of K were computed. (See table 3). 
The mean values of K are remarkably constant for the different quartiles, 
in fact all are within the range of their probable errors. This may be regarded 
as evidence that the growth constant has the same value for all classes of plants 
in this population without regard to their relative heights, since the relation 
between the final height and the height at any given time obeys the same 
principle. 
A brief consideration of a parallel case will make it evident that such a re- 
lation must exist in this sort of reaction. The inversion of cane sugar is a fa- 
miliar case of autocatalysis in which one of the products catalyzes the reaction. 
