BUREAU OF PLANT INDUSTRY. 415 



The usefulness of the method used for accurately determining the 

 length of staple of cotton has been demonstrated. The measurement 

 of samples from various lots of cotton with a view to selecting cotton 

 for use in the preparation of length types is still in progress. These 

 types will range in length from three-fourths of an inch up to If 

 inches, by eighths, and during the present year it is expected that 

 much further advance will have been made toward staple standards. 



Moisture in baled cotton. — All cotton handled in the investiga- 

 tions already referred to was carefully weighed when first baled and 

 at each subsequent handling, and the history of these bales is very in- 

 teresting. In general, it may be said that the bales from Texas gained 

 about 10 pounds in weight before they were finally spun. This cotton 

 was picked during dry weather in Waco and milled at intervals of 

 three to five months later in various sections of the country without 

 in the mean time having undergone ocean transportation. As the 

 bales were completely covered and never sampled, this gain of 10 

 pounds was purely one of absorption of moisture. As far as known, 

 this experiment affords the first systematic record of this kind. The 

 results are very significant and prove that a study of the moisture 

 content of cotton along these lines is very important. Here again 

 these results of a single season are to be supplemented by similar 

 observations to extend over several seasons and to include a greater 

 variety of conditions. 



RooT-ROT COTTON. — Advantage was taken of the circumstance that 

 root-rot was prevalent at Waco, Tex., last season to pick and handle a 

 bale of such cotton separately. Other bales of sound cotton were 

 taken, for comparison, from the same field. These bales were made 

 into yarn with the following results: (1) The grade of the root-rot 

 cotton is slightly lower, practically half a grade; (2) the loss in mill- 

 ing root-rot cotton is about 13 per cent as against 10 per cent, i. e., 30 

 per cent greater; (3) the breaking strength of yarn made from sound 

 cotton is greater, being approximately 94 pounds as against 86 

 pounds, i. e., about 10 per cent. 



These differences are sufficient to justify attempts at classifying 

 cotton in the field. There is no reason to doubt that the separation 

 of poor cotton from good cotton will be relatively just as profitable 

 as the separation of poor fruit from good fruit, or poor grain from 

 good gram, or poor wool from good wool. If this root-rot factor and 

 other factors which reduce the quality of cotton were better known 

 and the cotton classed in the field, the resulting bales of cotton would 

 be of such a character as to command a more profitable market. 



Ginning cotton. — Ginning experiments have been continued dur- 

 ing the year, and for the first time uniform samples were obtained in 

 sufficient (Quantity to feed to a commercial GO-saw gin. These samples 

 were obtained in sets of four by placing four locks from a boll of cot- 

 ton in four separate bags. The tests obtained from these samples are 

 open to none of certain important objections that cxin be brought 

 against the samples of seed cotton ordinarily used. The results of 

 the tests are in themselves interesting and valuable. The method has 

 proved so satisfactory that it is now being applied in an attempt to 

 define the factors which enter into the most successful ginning, as 

 well as the manner in which these factors affect the grading and 

 standardization. 



