155 



the total number of pupae secured is less than would have been 

 the case had the fruit been left to ripen on the tree and fall nor- 

 mally, thus giving the parasites opportunity to attack as they 

 developed those larvae that were too small for them at the time the 

 fruit was gathered. The writers secured larvae at the insectarj 

 as they emerged from the fruits daily. It is evident, therefore, 

 that, as larvae do not generally emerge from fruits until they are 

 full grown, the last larvae to be secured from any lot of fruit 

 would show no parasitism, or a parasitism very much less than 

 would the larvae emerging during the first few days after the 

 gathering of the fruit. This point is made clear by the data in 

 Table I, secured from larvae emerging from coffee cherries picked 

 in Kona, Hawaii. 



Table I. Parasitism among larvae developing in coffee 

 grown in Kona, Hawaii. 



Coffee grown Date of Lai^'ae emerged Percentage of lar- Total per- 



at 



Kahului 



collection 

 of sample 



Oct. 31 



between 



vae parasitised centage of 



Hookena 



Oct. .31 



Kahaloa 



Oct. 30 



Keauhou 



Oct. 30 











by 



parasitism 











DtUS 



D. 













humilis 



tryoni 





Oct. 



31- 



— Nov. 



3 



31.2 



2i.8 



53.0 



Nov. 





-Nov. 



6 



20. 



12.3 



32.3 



Nov. 



6- 



-Nov. 



9 



5.3 



.7 



6.0 



Nov. 



9- 



-Nov. 



11 



0.0 



.0 



0.0 



Oct. 



31- 



-Nov. 



3 



16.7 



40.7 



57.4 



Nov. 



3- 



-Nov. 



6 



17.0 



19.0 



36.0 



Nov. 



6- 



-Nov. 



9 



1.2 



0.0 



1.2 



Nov. 



9- 



-Nov. 



11 



0.0 



0.0 



0.0 



Oct. 



30- 



-Nov. 



8 



80.4 



2.2 



82.6 



Nov. 



3- 



-Nov. 



6 



23.3 



0.4 



23.7 



Nov. 



6- 



-Nov. 



9 



3.7 



0.0 



3.7 



Nov. 



9- 



-Nov. 



11 



0.0 



0.0 



0.0 



Oct. 



30- 



-Nov. 



3 



85.2 



0.0 



85. 2 



Nov. 



o 

 ■J— 



-Nov. 



7 



55.2 



0.0 



55.2 



Nov. 



7- 



-Nov. 



9 



6.7 



0.0 



6.7 



Nov. 



9- 



-Nov. 



11 



3.7 



0.0 



3.7 



The data in Table I leave no doubt that only those percentages 

 of parasitism secured from larvae emerging soon after the fruit is 

 gathered represent the real effectiveness of the parasites for any 

 particular lot of fruit. And it is because this fact was not fnlly 

 appreciated by the writers until the above data w^ere secured that 

 they are not able to state dc^finitely that each percentage I'ecordcd 

 in Tables II. -Y, is based upon the first lot of larvae secured from 

 each lot of fruit. Often si^mples of fruit were taken at frequent 

 intervals from the same locality, hence elimination of all but first 



