20 



CIRCULAR NO. 128, BUREAU OF PLANT INDUSTRY. 



impossible to select single strands for comparison. The hard filjers, 

 however, present single strands sufficiently uniform and free from 

 other strands to be selected for breaking, but they are not uniform 

 enough to be directly compared, A coarse, hard fiber may break at 

 a higher tensile strain than a finer one, yet the fine one may be much 

 stronger when compared with equal bulk or weight for weight. 

 Weight is more satisfactory for comparison than diameter or bulk 

 and it is more easily determined. For comparison, therefore, the 

 breaking strain of each sample is computed to the conA^enient but 

 arbitrary standard of a fiber weighing 1 gram per meter of length. 



Fig. 2.— Testing machmc with a capacity of 1 gram to 90 kilograms, used for determining the strength of 



hard fibers. 



In testing hard fibers the following method is pursued: 

 Twenty fibers are selected from the sample and their length meas- 

 ured, after which they are weighed and the total weight is di^aded by 

 the total length, giving the average weight per meter. They are then 

 broken, one at a time, in the testing macliine. (Fig. 2.) The average 

 breaking strain divided by the average weight per meter gives the 

 breaking strain per gram meter. 



The results of numerous tests of some of the principal hard fibei-s 

 are given in Table III. 



H'ir. l:.'8] 



