SPINNING TESTS OF UPLAND LONG-STAPLE COTTONS. 11 
It is noteworthy that the actual waste obtained was in every case 
decidedly below the usual mill estimate for this wider setting. The 
Lewis bale again showed the Least waste, while the Columbia and 
the Delta lost identical percentages. 
Averaging the two Webber bales we have the following showing 
for the five varieties as compared with the arbitrary commercial 
or standard expectation of 24 per cent of loss at this setting. The 
order of excellence then appears as follows : The Lewis showed an 
actual loss of 11.85 per cent less than the standard. Columbia and 
Delta each showed 8.85 per cent less. Webber showed 4.04 per cent 
and Durango 3.57 per cent less, respectively. 
Compared geographically the Texas bale showed a total waste on 
the wide setting of 20.43 per cent; the Delta, five bales mixed, lost 
16.15 per cent; the Carolina staples, four bales averaged, lost 17.05 
per cent. Special attention is called to the fact that the bale of 
Columbia which exactly equaled the Delta in this test was one- 
sixteenth inch shorter cotton than any of the others, save Durango, 
and must have possessed superior uniformity, else it would have lost 
more than the longer staples. 
The very slight superiority shoAvn b}^ the Delta over some of the 
Carolina staples at this wider setting does not offset the much greater 
superiority of the Carolinas over the Deltas at the 13 to 15 per cent 
setting, the latter being more frequently used in actual millwork. 
NUMBER AND UNIFORMITY OF TESTS. 
A study of the large number of percentage tests and weighings 
made on each variety, as shown in Table VI, ranging from 17 on 
Lewis to 36 on Webber No. 1, shows conclusively that at no time 
during the run on any bale was the variation sufficient to cause any 
uncertainty as to the amount of short fiber contained in the sample. 
The widest variation between any two weighings on the same variety 
is found in the Durango, namely. 1.48 per cent. In the Delta the 
widest variation was 1.43 per cent. Even under the' most ideal con- 
ditions the machinery builders never claim for their combers less 
than about 2 per cent variation for different runs on the same cotton, 
even on the best Peelers and Egyptians. It appears, therefore, that 
each of the varieties tested was practically uniform throughout the 
entire run made upon it, such variation as occurred being within 
the limits of error of the machine. The weighings or tests were 
made at intervals of 10 to 20 minutes during a two to three day run 
on each bale. 
