290 
Journal of Agricultural Research 
Vol. XIV, No. 8 
Table II shows what a large variation may occur in the averages of the 
breaking strengths of ioo fibers. In the case of No. 991, the fibers are 
fairly uniform, and there is less variation. The variations are so great 
in most cases that one would not be justified in making any final deduc¬ 
tions from the results. The differences which occur in the breaking 
strengths of different fibers in the same sample may be more clearly seen 
by comparing the results of humidities 40 and 70 in Table II with similar 
results upon the same sample as shown in Table I. Such large variations 
will occur when the size of each fiber is left out of consideration. 
Table III shows the average diameter, breaking strength, and tensile 
strength for 100 fibers and the average tensile strength for 200 fibers. 
It can readily be seen that when the diameters of the individual fibers 
are taken into consideration there is much more uniformity in the re¬ 
sults obtained. Had 500 fibers been tested, it is no doubt true that the 
average tensile strength would have been somewhat more accurate than 
when only 200 fibers were tested. In the case of sample 991 (humidity 
70) 600 fibers were broken, and the average tensile strengths for each 
