804 
Journal of Agricultural Research 
Vol. XXI, No. XI 
In the work at hand, instead of a rapid increase in the early stages and 
then a falling off in size of increments, growth comes about by con¬ 
stantly increasing increments. It did not appear in the present study 
that the data were amenable to a logarithmic type of curve. 
Blackman 1 has applied the coiripound interest law, commonly used 
by physicists, to plant growth, particularly to changes in weight under¬ 
gone by annual plants from the germination stage to more mature 
stages. A fuller discussion of this law will be undertaken when the 
data have been presented. 
.« ; i ; CVIM FORMATION IN BROMUS INERMIS 
In the spring of igiida large number of seedlings of Bromus inermis 
Leyss were started in flats and later planted out in the field as individual 
plants 3.5 feet apart each Way. 'The' 6e£d came from a Canadian source 
and was commercial seed. It had not been pedigreed or selected to any 
type. On June 19 countings were begun of the number of culms per 
plant of a ; limited number of plants. 2 At the tinie of first counting the 
plants had 'a height of 92.2 trim/ as ari average of the longest culm per 
plant, measuring to the extreme tip of the leaf. Stooling had begun on 
most of the plants. This consists essentially, in its earlier phases, in the 
formation and growth of additional culms springing from, at, or near the 
base of the primary culm. This study is concerned primarily with the 
rate of production of such secondary culms and the relation of their 
number to other characters of the plant. 
Countings were made every five days except for the intervention of 
Sundays and holidays. The dates of the countings were June 19, 24, 29; 
July 5 > IO > 2 °y 25, 31; August 5, 10, 15, 22, 26, 31; September 5, 11. 
There were thus a total of 17 countings, embracing a period of 85 days, 
including the dates of the first and last countings. At the time of the 
first counting the number of culms per plant 3 * averaged 2.37, with a 
minimum of 1 and a maximum of 5. It is seen that at the first count 
growth was still in its earlier stages, from the standpoint of this study. 
At the date of the last count, September 11, culm formation was still in 
progress and apparently unaffected by weather conditions. The count¬ 
ings were stopped at this time, primarily because the large number of 
culms per plant made a precise count difficult. 
The countings were continued over an interval of time sufficiently long 
so that the law of their rate of formation during favorable growth con¬ 
ditions can be deduced with a reasonable degree of satisfaction. A study 
of the rate of cessation of formation during the close of the growing 
1 Blackman, V. H. the compound interest law and plant growth. In Atm. Bot., v. 33,110. 131 
P- 353-360. 1919. ’ 
2 Data were secured on 97 plants for the period of study. 
3 A culm was recognized as such when it had attained an extreme height of 25 mm., measurement being 
made from the tops of stakes driven down nearly flush with the surface of the ground. 
