11 



On Table YlU'the figures obtained on Tables IV, V, VI, and VII 

 are re-arranged to show the distribution around the average of the 

 numbers of attacked bolls. It will be seen that it is a one to three chance 

 for any sample being within two places above or below the average, a 

 three to two chance that it will lie within five places above or below the 

 average, and a three to one chance that it will lie within the seven 

 places above or below the average. 



An attempt was made te estimate the influence of the neighbour- 

 hood to the cotton fields of (1) a ginnery, (2) a large village, samples 

 being taken north, east, south, and west, in the immediate vicinity and 

 also at a distance. The figures obtained were neither significant nor 

 consistent and we refrain from publishing them. 



The figures obtained this year will be of special use as material 

 for comparison in future years, and especially they should afford infor- 

 mation at any early date as to the prospects of damage to the cotton 

 crop. They will also help to show whether the methods employed to 

 cope with Gelechia are giving the results hoped of them, and whether 

 the methods have been applied equally well everywhere. 



This year's work has been of additional value in indicating certain 

 modifications which will be necessary in carrying out similar work in 

 the future. Chiefly it is apparent that if weekly figures for each 

 Markaz are desired, larger samples will have to be taken, and that 

 about 1,000 bolls weekly will be required from each locality. The 

 method of taking a few bolls here and a few there throughout each 

 district ought to be adhered to, in order to obtain really average results. 



Having shown in the foregoing the method of obtaining the 

 samples, and the value of a single observation, it now remains to see 

 what general conclusions can be drawn from the material available. 



On account of the dissimilarity of conditions in the Delta and 

 Middle Egypt as compared to Upper Egypt, it is proposed in the fol- 

 lowing to leave Upper Egypt unconsidered and to confine ourselves to 

 the examination of the mean curve for the Delta and Middle Egypt. 

 (See Diagram.) 



Disregarding small irregularities, it will be seen that the percen- 

 tage of bolls attacked was fairly stationary during the whole of July. 

 Increase began to be evident in the first week of August, and the 

 rate of increase was almost uniform until the second week of October, 

 when the maximum was reached. This curve chiefly shows how 

 quickly Gelechia can utilize the available food supply. 



