96 STUDIES IN GUJARAT COTTONS 
as compared with any others, and they seem, moreover, to have a somewhat 
different effect in different strains ( e.g., A26and B3). But there are strains 
which consistently give a very high ginning percentage like E5, and which may 
have great importance from a breeding point of view on this account. 
A critical examination of these figures, however, suggests that the ginning 
percentage is itself a complex, being dependent on the weight both of the 
seeds and of the lint. A rise in the ginning percentage might in fact be brought 
about either by an increase in the lint weight or by a decrease in the seed 
weight, and hence a better figure to measure the proportion of lint ina well- 
developed boll would be to compare the weight of lint with the number of 
seeds and not with their weight. That is to say, we should use a figure which 
compares the weight of lint per 100 seeds. This figure has been already used 
by the writers in Egypt under the name “ Lint-index ” and its value in the 
cases quoted in the last table is as follows :— 
Lint-index ( weight of lint in grammes per 100 seeds ). 
Grm. 
Strain A26.. eee LOUG—lh7. 4°31 
1917-18 ' 4-29 
1918-19 4°80, 4°91 and 5:04 
1919-20 4°82 
Strain B3 sh .. 1916-17 4-11 
1917-18 4:04, 4°13, 4:05 and 4°25 
1918-19 4°66 and 5°05 
1919-20 5°15 
Sirain B21) ek -- 1916-17 4°01 
1917-18 3°98 
1918-19 4°30 and 4°30 
1919-20 4°60 
Strain C22 ueaer. -.. 1916-17 4°33 
1917-18 3°73, 3°93, 3°92 and 3°95 
1918-19 4°29 and 4°37 
1919-20 4°80 
Strain E5 5 .. 1916-17 4°49 
1917-18 4°75, 4°69 and 4°84 
1918-19 5°50 and 4°96 
1919-20 o%3 
The constancy of the lint-index is greater than that of the ginning 
percentage as is well illustrated in the case of Strain A26, 1919-20, as against 
1918-19, and hence it appears to be a better factor to use in considering the 
relationship of the lint and the seed. 
It is a general belief among cotton merchants that variation in the 
ginning percentage in one type of cotton is due simply to variation in the seed 
weight. Our results show, however, that this is not the case, and that in any of 
our pure strains the variation in the seed weight is only one of the factors 
