296 



DR. L. H. GOUGH. 



If it be permitted to put forward a theory to explain the figures, it may be 

 supposed that the actual loss of seeds occurs only in such bolls as are attacked at 

 an early stage of their development, and that the loss in numbers is actually due to 

 entire seeds having been eaten up, shell and all. Loss of seeds will consequently 

 be greater when the bolls are attacked at an earlier stage. Very young bolls are 

 usually not infested, until the attack has become more or less general. The average 

 number of seeds produced per 100 bolls can consequently be taken as an index to 

 the severity of the attack in the same way as seed weight of sound seeds, or germina- 

 tion of sound seeds. 



Table XII. To show Correlation between the Intensity of Gelechia Attach and 

 Reduction in the Number of Seeds found per 100 Ripe Bolls of Sakellarides, 



1917 Crop. 



Percentage 



Number of 



Average number 



Total number 



of sound 



bolls 



of seeds per 



of seeds in 



seeds. 



used. 



100 bolls. 



sample. 



100 



180 



1,637 



2,946 



94 



1,723 



1,509 



25,991 



81 



1,341 



1,459 



19,562 



74 



4,404 



1,335 



58,790 



68 



686 

 98 samples 



1,473 



10,107 



Weighting as in Table VII : r = 0-889 ± 0-063. 



I) 



The Effect of Gelechia Attack on the Amount of Lint produced. 



The effect of Gelechia attack on the amount of lint produced is more difficult to 

 estimate than the damage done to the seed by the same cause. This is chiefly due 

 to the lint becoming inseparably mixed during ginning. For this reason we have 

 to rely on indirect information in estimating loss of lint. The amount of lint produced 

 per sound boll varies very considerably, the quantity depending largely both on the 

 number of seeds produced and on their weight ; the amount produced per seed is to 

 a very great extent proportionate to the weight of the seed. 



Any factor that can affect a developing seed can also affect the quantity of lint 

 produced by that seed. Thus it is obvious that Gelechia attach must affect the amount 

 of lint produced (I) by reducing the number of seeds matured per boll, (2) by reducing 

 the average weight of the sound seeds produced in attacked bolls, and (3) by disturbing 

 the development of immature attacked seeds. 



The loss of lint-weight can consequently to some extent be calculated from the 

 damage done to the seed, provided that the proportion lint-weight to seed-weight 

 is not seriously upset. As it is more usual to use the " percentage lint " than the 

 proportion of lint-weight to seed-weight, that method of expressing the proportion 

 of lint in a given weight of seed cotton will be used in the following discussion. 



Before considering the effect of Gelechia attack on the " percentage lint," some 

 general observations on factors which cause the percentage to fluctuate in normal 

 seed cotton require notice. 



