3 



This method of calculation cannot however be applied without 

 modifications being introduced to allow for the obvious fact that 

 each boll can only harbour a limited number of worms, only a certain 

 amount of food and space being available. 



The number of seeds set in sound bolls of Sakellarides were found to be 

 on the average in two different samples of 180 and 100 sound bolls 16.37 

 (Gough,Bull.Ent.Res. 1919, page 296), or 14-99 (page 290) showing that 

 the number of seeds set per lock varies between five and six, or fifteen to 

 sixteen per boll, fifteen being in the two samples examined more frequent. 



The number of worms developing completely in a boll depends 

 however as already stated on the amount of food offered. The follow- 

 ing figures may facilitate comprehension. 



The weight of 100 mature sound seeds in a sample of Sakellarides 

 cotton was found to average 10-300 grammes, 1,000 seeds being used. 



Thirty sets of 100 double seeds from the same sample were weighed 

 (without worms), the red and black components being weighed sepa- 

 rately. The weight of 100 black components was found tQ bQ^-j92ft 

 grammes, the weight of 100 red components 5-578 gramme^,. 3,000 

 seeds being used in each case. The damage done by 100 "worms.; to 

 200 nearly ripe seeds was consequently 8-097 grammes, or 79 per 

 cent of a seed per worm by weight. The damage done to younger 

 seeds would necessarily be much greater, corresponding to their less 

 weight and greater moisture content. 



One boll of Sakellarides could consequently raise a maximum of 

 twenty worms to maturity, if all seeds were completely destroyed 

 and if the attack was delayed until the seed was nearly ripe. 



It may be supposed that ten worms are the average maximum 

 number likely to develop in one boll. This does not mean to say 

 that no boll will contain more than ten, twelve having actually been 

 observed by us; it also does not infer that the limiting factor never 

 begins to act until ten have attacked the boll. But as an average 

 number ten seems to be not far out. If anything, it probably errs on 

 the side of excess. 



The question instantly arises, What is the limiting factor which 

 prevents more than a certain number of worms occurring simulta- 

 neously in a boll? There are three possibilities which suggest them- 

 selves at once, which give different results when calculating the total 

 number of worms required to produce any percentage of infestation, 

 but wlrch do not alter the probable number present in samples of 

 open cotton bolls such as would normally be picked for crop. They 

 are : 



( 1 ) The number of worms may be li mited by the attacked boll drying 

 up or falling off the plant, when more than a certain number of worms 

 have attacked it. This would divide bolls produced into three classes : 



