98 THE AMERICAN MONTHLY [May, 



ginned in proportion to the distance from the surface reached by the 

 cutting edges of the teeth of the ginning saws. The thread or fibre in 

 its young state is cylindrical, but upon maturing and becoming dry it 

 collapses and assumes a more or less flat, ribbon-like, twisted form. 

 The degree of twist given the fibre, its regularity in diameter and 

 length determine the value of the cotton in the markets of New York 

 and Liverpool. 



III. Varieties Tested. — I obtained a number of samples of cot- 

 ton, representing eighteen varieties, two selected specimens from 

 Savannah, Ga., of the vSea Island cotton and a sample of the '' Bailey " 

 fibre from North Carolina. Careful sti'idies have been made of these 

 specimens under high powers of the microscope, and a number of in- 

 teresting results were obtained. The following are some of the qnes- 

 tions considered : 



I. How many real varieties of cotton exist? 



3. In forcing the plant under high cultivation is the fibre improved, 

 or is simply the '' weed" enlarged to the detriment of the staple.'* Is 

 it not often the case that the fruit of the cotton plant is damaged by too 

 rapid maturing, just as the fruit of the peach is known to be immature 

 at the centre in some early forced varieties .'' 



Many experiments were made on each of the samples to determine 

 the diameter and regularity of fibre, the average length of the strands, 

 the character of twist and the internal structure. Also several strands 

 were selected at random from the bolls and the strain necessary to 

 break them carefully determined by fastening one end of the fibre and 

 weighting down the other until rupture was produced. 



Sample number i was a poor grade of cotton that was obtained from 

 stalks about ten to twelve inches high, growing on sandy soil unfertil- 

 ized. Four tests of the strength were made with two strands in each 

 test with the following results: ist. Broke under a strain of 9.498 

 grammes; 2d, 19057 grammes; 3d. 31.404 grammes; 4th, 11.635 

 grammes. Average for two strands 15.398 grammes (r gramme is 

 equivalent to 15.43 gi'ai"s). Length of fibre 1st test, 22.4 millimeters 

 (i millimeter is equivalent to 0.039 of an inch); 3d test, 34 millime- 

 ters; 3d test, 23.3 millimeters; 4th test, 24.8 millimeters. Average 

 length of fibre, 33.6- millimeters. The diameter ranged from 0.009 

 millimeters to 0.016 millimeters. These results indicate a lack of uni- 

 formity. When the fibres were placed under the microscope it was 

 noticed that some were immature, some were only slightly twisted, 

 while others, though well twisted, were small and weak. 



Number 3 was obtained from the same field but from a plat that had 

 been fertilized with floats. The stalk was small and badly blighted, 

 but the fibre was about the same grade with number i . The grade of 

 both plants is so low in the scale, the injury to the fibre may be due 

 largely to immature growth in the plant before the blight obtained 

 headvvay. However, this is difficult to determine with these samples, 

 as they were obtained so late in the season. 



Number 3 was also obtained from the same field and the same plat 

 from which No. i was drawn. The plant was not blighted, but the 

 stalk was small. In the first test upon the strength of two strands the 

 fibre broke under a load of 27.475 grammes ; 3d, 33.915 grammes ; 3d, 

 29.000 grammes; 4th, 45.176 grammes. Average for two strands, 



