16 BULLETIN 121, U. S. DEPARTMENT OF AGRICULTURE. 
After the cotton has reached the mill it is still possible to get 
varying results from the same bale by running the beaters too fast, 
thus breaking the fibers, by feeding too heavily, or by operating with 
improper adjustments of beater grids, or by allowing too strong an 
air current on the pickers. 
On the card it is possible to get a wide variation in results, even 
when the machine is supposed to be adjusted accurately. The reason 
for this is that the most important settings or adjustments are made 
anywhere between five and fifteen one-thousandths of an inch. A 
difference of one or two thousandths will measurably affect the per 
cent of waste. It is therefore apparent that only the most painstak- 
ing care can obviate these possibilities of error. Again, if the speed 
of certain parts of the card is varied, or if a heavier lap is fed. or 
if the grinding of the steel wire clothing of the cylinder, doffer. or 
revolving flats is neglected, or if the stripping of these is too infre- 
quent, a varying amount of waste will result. 
At the comber a number of conditions may cause a variation in 
the amount of waste. Among these are differences in speed, in the 
angle of the top comb, in the distance between the nippers and the 
detaching rolls, in the wefght of the lap, in the rate of feed, the 
timing of the parts, etc. If any of these factors are changed, even 
in a slight degree, a different result is obtained. The temperature 
and humidity also affect the results of such tests. 
Items enough have been enumerated to show the possibility of 
error in tests of this nature. However, as all of these tests were 
made on the same machines, and all except the Delta, at the same 
time, and without the slightest mechanical change, the results ob- 
ianed are strictly comparable as to five samples and very closely 
comparable as between these and the Delta. 
TESTS OF BREAKING STRENGTH. 
Table X shows the breaking strength in pounds per skein of the 
carded and combed yarns. All varieties wore spun into 50s yarn, and 
the twist per inch was calculated as follows: The standard twist 
factor for filling yarns being 3.25, this factor, multiplied by the 
square root of the counts (50s) gives the twist per inch thus: \/50= 
7.07; 7.07X3.25=22.97 turns per inch. 
In preparing the yarn for these breaking strength tests it is tha 
custom of the trade to reel off a skein containing 120 yards. The 
reel is 54 inches in circumference, and when the skein is attached to 
the testing machine it is an endless length of yarn. This skein is 
hung upon two hooks and when the tester is started these hooks begin 
to draw slowly apart. The strength of the skein is registered on a 
dial; 
