286 
Journal of Agricultural Research 
Vol. XIV, No. 8 
Lewis (5) made tests on woolen and worsted yams, similar to those of 
Barker and his coworkers, under controlled conditions of temperature 
and humidity, at five different humidities ranging from 45 to 85 per cent. 
He found an increase of 16 per cent in the tensile strength of cotton and 
a decrease of 18 per cent in the tensile strength of worsted for a rise of 
40 per cent in the relative humidity. 
The work carried on at the Wyoming Experiment Station in 1911 
under the direction of Hill (4) showed that the dry wool fiber was stronger 
than the wet fiber, and that at a humidity of approximately 15 per cent 
the wool fiber was stronger than at 35 per cent. Because of the lack of 
the means of temperature and humidity control, this work was tem¬ 
porarily suspended until such control conditions might be established. 
EXPERIMENTAL WORK 
On undertaking research studies upon wool the writer found that 
it was first necessary to improve further the means of measuring the 
strength of the wool fiber before a continuation of studies in the effects 
of chemical reagents and of alkali and weathering could be made with 
satisfactory results. In September, 1917, the writer succeeded in 
bringing a small inside room under automatically controlled conditions 
of temperature and humidity. A description of this room will be found 
elsewhere in this article. The work of Hill (4) who tested over 59,000 
fibers, clearly showed that it was quite impossible to get satisfactory 
results by testing the single wool fibers under ordinary room conditions. 
He states (p. 123): 
The variation of the means of hundreds is so great that the mean of this or a smaller 
number of tests is a very inaccurate measure of the mean of a sample of wool containing 
only a few thousand fibers, and that the means of thousands can scarcely be used for 
anything more than the most general work. 
Anyone who has tested textile fibers knows that to test only 500 wool 
fibers is not only a long but a tedious operation, and it would be impracti¬ 
cable to test many samples, were so many tests required for each sample. 
It was thought, however, that possibly under controlled conditions of 
temperature and humidity the number of fibers necessary to be tested 
on each sample, with satisfactory results, might be greatly reduced. 
With this thought in mind the writer began the work covered in this 
paper, with a plan outlined to test samples of wool fibers at five humidi¬ 
ties ranging from 40 to 80 per cent. Samples of wool from the shoulders 
of four sheep, a Rambouillet, an Oxford, a Cotswold, and a Dorset were 
selected. All tests were made upon single fibers from locks of wool 
which had not been cleaned or scoured. The tests were all made on a 
Reeser and Mackenzie fiber-testhig machine, a machine devised by 
Matthews, of the Philadelphia Textile School, and fully described in 
Matthews’s Textile Fibers (6, p . 234). 
