8 i6 
Journal of Agricultural Research 
Vol. XXI, No. II 
From Table IV one notes a distinct rise in the indexes up to the second 
or third counting period and then a gradual but somewhat irregular falling 
to the last period. The indexes as a whole decrease in size as the initial 
culmage increases. For the group of plants initialing at 4 the efficiency 
index, or rate of interest, averages less and is uniformly less than for any 
of the other groups. During the early stages the rate of increase is con¬ 
siderably higher with the plants with fewer culms. An individual plant 
with 1 culm doubles its culmage at the first increase, but with a 4-culmed 
plant the culmage need be increased but by 25 per cent. It would seem 
to appear in this case that the plants with the larger initial culmage are 
less “efficient” than the smaller initialing group and that this diminished 
“efficiency” is automatically due to the larger size of plant. This con¬ 
dition is quite different in those cases cited by Blackman, where the small 
initial seed weights lead directly to a lessened efficiency index. It is not 
evident that the indexes dealing with culm formation in brome grass have 
any real organic significance. 
SUMMARY 
A study is made of fate of culm formation in Bromus inermis. The 
rate of increase is not fixed but takes place at an accelerating rate. An 
equation was developed which expresses the rate of increase in a sym¬ 
bolical manner. The resulting curve is of the parabolic type. 
An analysis of the data shows that plants initialing at culmage 1, 2, 3, 
and 4 become, on the whole, increasingly more divergent in number of 
culms during the season. This increasing divergence may be explained 
from the initial culmage and the consequent mechanics of further culm 
formation. It is not evident that the compound interest law has any real 
organic significance when applied to culm formation in Bromus inermis. 
