44 Ginnhif/ Percent of Indian Cottons 



boll. The sample is gathered after the capsule has expanded thoroughly 

 and when the lint and seed are thoroughly drj'. From this sample fibres 

 to the number of about 2000 are accurately counted and weighed. The 

 remaining lint is removed from the seed by a small hand gin and the 

 weight of lint and seed recorded as is also the number of seeds. The 

 volume and specific gravity of the seed are then determined by displace- 

 ment in water. The data thus obtained give 



(1) weight of a known number of fibres, 



(2) weight of total fibre, 



(3) weight of seed, 



(4) number of seeds, 



(5) volume of seed, 



from which the figures required can be derived by direct calculation. 



The method is not ideal as, apart from the precautions necessary in 

 the process of sampling, a single complete determination, with subsequent 

 calculation, occupies at least two hours — a point of considerable import- 

 ance where the value of the results is to a large extent determined by 

 the number of the observations. 



During the course of the investigation it soon became apparent that 

 the specific gravity of the seed, even within the limits of a single sample, 

 was subject to marked fluctuation. Evidence, as far as it was obtain- 

 able, indicated, however, that such fluctuation depends on the conditions 

 which prevail during the process of ripening. Any inherent dift'erence 

 in specific gravity due to the nature of the plant is small in comparison 

 with such observed fluctuation. It has been considered advisable, 

 therefore, to eliminate this character by reducing the specific gravity 

 in all cases to an uniform figure of I'lO, such a correction involves a 

 coiTesj)onding correction of the observed giiming percent. There thus 

 remain three characters, volume of seed, number of fibres arising from a 

 single .seed and weight of the individual fibres, which, from the given 

 a priori considerations, there is reason to suppose might influence the 

 ginning percent. The object of the present work is to determine how 

 far fluctuations in the ginning percent can be accounted foi- by these 

 characters only and whether a search for other characters must be 

 undertaken. 



The units used for the expression of these values are 

 volume of seed ... cubic millimetres. 



No. of fibres per seed ... in thousands, 

 fibre weight ... weiaht «if 1000 fibres in milliorams. 



