EFFECTS PRODUCED BY PINK BOLLWOBM ON THE YIELD OF COTTON SEED. 291 



rises. Irregularities only occur when the samples combined to obtain an 

 average have been reduced to too small a number ; this is doubtless due to the 

 resultant greater chance of including non-representative samples and for changes 

 in humidity to make themselves noticed in such cases. The correlation between 

 fall in the percentage of sound seed in a sample (increase of Gelechia attack) and 

 fall in the average weight of a given number of sound seed is very high, having 

 been found to be - 976 (± - 008), leaving no possible doubt as to the connection 

 between the two sets of observations. 



Table VIII. Showing Number of Seeds of each - 01 gramme difference in 

 iveight in sound and attacked samples. 









Intensity of attack. 





Weight in mg. 



Sound 

 samples. 























8 per cent. 



21 per cent. 



34 per cent. 



46 per cent. 



140-150 







2 







130-140 



168 



76 



6 



11 



— 



120-130 



151 



92 



14 



31 



1 



110-120 



152 



360 



19 



■ — 



8 



100-110 



60 



159 



33 



19 



14 



90-100 



40 



62 



25 



8 



17 



80-90 



25 



10 



46 



71 



29 



70-80 



20 



12 



28 



188 



57 



60-70 



14 



5 



38 



18 



24 



50-60 



6 



9 



21 



9 



31 



40-50 



7 



1 



8 



39 



10 



30-40 



5 



2 



5 



16 



4 



20-30 



1 



— . 



— 



10 



1 



10-20 



— 



— 



— 



10 



1 



The same as above, but with the Numbers of Seeds replaced by the Percentage for the 

 Sample for convenience in comparison. 



— i 1 



140-150 



130-140 



120-130 



110-120 



100-110 



90-100 



80-90 



70-80 



60-70 



50-60 



40-50 



30-40 



20-30 



10-20 



26 

 23 

 23 



9 

 6 

 4 

 3 



2 

 1 

 1 

 1 

 



10 



12 



46 



20 



8 



1 



2 



1 



1 











2 



6 



8 



13 



10 



19 



11 



16 



9 



3 



2 



3 



7 



4 



2 

 17 

 44 



4 

 2 

 9 

 4 



2 

 2 



1 



4 



7 



9 



15 



29 



12 



16 



5 



2 



1 



1 



However, on comparing this table with Table II, in which the average weight 

 of many samples of seed are recorded, it will be observed that there the weight 

 of the seed is not so evidently influenced by the severity of the attack. This is 

 due to sampling, the material used for the two sets of samples not having received 

 similar previous treatment. The Ministry's selected seed was of course all ginned 

 in commercial ginneries, and was sifted in the usual course of events. During this 

 sifting all small seeds and most of the broken fragments are removed. The seed 



(C541) b2 



