Sept. i, X99x 
Rate of Culm Formation in Bromus inermis 
81 3 
From tlie foregoing explanation, it is seen that only the peripheral 
culms produce sister culms, and at any stage of increase a peripheral 
culm produces only one sister culm. The series resulting when the 
initial culmage is 4 is obviously as follows: 4, 8, 16, 28, 44, 64, etc., the 
first difference being continually augmented by 4. The general expres¬ 
sion for this series is 
y=23? — 2#+ 4. 
Now, the point in question is whether the results obtained in the 
field agree reasonably well with the figures obtained from the conven¬ 
tional plants. In the first place, one is dealing with the phenomenon 
of growth, which develops in a continuous manner. However, since 
the new growth is not recognized until it attains a certain stage of develop¬ 
ment in a culm of certain height, the accretions may be considered as 
being discrete in character. The method of recording the accretions, 
by counting, was necessarily discrete. The time elapsing between was 
chosen in a purely arbitrary manner, and in any explanation or symboli¬ 
cal description of rate of culm formation it would be purely a matter 
of choice if the length of time between countings were modified in one 
way or another. If the time periods were to be arbitrarily reduced 
from those used in the field, the number of culms appearing at each 
counting would have to be interpolated in so far as the new periods 
were concerned. The same would hold true if the time periods were 
increased unless, indeed, the new time periods were made some multiple 
of those actually used. As a matter of convenience, it seems better to 
double the period length between countings when comparing conven¬ 
tional plants with the actual ones. The number of countings would 
therefore be reduced to nine with an interval of about 10 days between 
successive counts. 
In Table III are given the counts made upon the four series of plants 
based upon 10-day intervals, and for comparison there are presented 
tlie figures from the four conventional plants having a rate of culm 
increase according to the scheme outlined. 
Table III. Comparison in rate of culm formation between four groups of plants and 
four conventional plants 
Number of culms. 
Date. 
A 
B 
A 
B 
A 
B 
A 
B 
June 19.... 
29. . . . 
July 10.. . . 
20.. . . 
a 
Aug. 10. . . . 
22.. . . 
31.. .. 
Sept. 11... . 
I 
3- 46 
6. 71 
9. 71 
20. 29 
3i- 50 
48. 14 
63. 86 
90. 04 
1 
2 
4 
8 
16 
28 
44 
64 
88 
2 
5. 08 
9- 67 
15. OO 
29. 42 
39- 63 
62.83 
83-63 
107. 42 
2 
4 
8 
16 
28 
44 
64 
88 
116 
3 
6.18 
14.46 
19. 96 
35- 79 
52. 29 
8 3- 7i 
IO9. 07 
143- 04 
3 
6 
12 
22 
36 
54 
76 
102 
132 
4 
5- 87 
*5- 53 
2I - 93 
39- 80 
58. 27 
86. 60 
109. 60 
140. 40 
4 
8 
16 
28 
44 
64 
88 
116 
148 
54131°—21-2 
