MEDITERRANEAN FRUIT FLY IN HAWAII. 99 



will take place in the warmer littoral regions about Honolulu where 

 checks due to cool weather are not so effective as in Kona. A study 

 of the data in Table XXVIII shows that Diachasma fullawayi gives 

 promise of being a most efficient parasite, particularly of fruit-fly 

 larvae in coffee cherries, since almost unaided it produced a mortality, 

 in one instance, of 92.5 per cent of 256 larvae emerging from coffee 

 cherries in Honolulu on December 8 to 10, 1915, which date was about 

 one year after it had been liberated. 



Data more recently secured by the junior writer indicate that instead 

 of supplementing the work of the opiines, Tetrastichus giffardianus will 

 prove a competitor, as its larvae appear to be able to hold their own 

 against opiine larvae within the same host, usually causing their death. 

 While both Tetrastichus giffardianus and Pachycrepoideus dubius (PI. 

 XX, fig. 2) have been reared during 1915 and 1916, neither is suffi- 

 ciently abundant to become an effective factor in control at present. 



GENERAL EFFECTIVENESS OF PARASITE CONTROL. 



Only a beginning has been made in determining the effectiveness 

 of parasites as a control factor in the Hawaiian Islands, yet the 

 rapidity of establishment and increase of the parasites has been very 

 gratifying. The data already published recording the percentage 

 of parasitism secured during 1914 and 1915, together with the addi- 

 tional data of Table XXVIII, indicate, however, that while para- 

 sitism in such fruits as the coffee cherry is remarkably high, in fruits 

 with a thicker pulp, such as the orange, it is very low. The data of 

 Table XXVIII have been chosen particularly as they demonstrate 

 that immense numbers of adult Geratitis capitata are developed in 

 spite of the excellent work of the parasites in certain host fruits. 

 Since adult fruit flies can live many months and oviposit quite regu- 

 larly as shown by the biological data, they have been able, with the aid 

 of the unprecedented variety and abundance of host fruits growing in 

 Hawaii, thus far to keep such an ascendency over their parasites 

 that they cause the infestation of practically all fruits which are per- 

 mitted to ripen. It would appear that unless effective pupal or egg 

 parasites are introduced, or unless care is given to the elimination of 

 host fruits which more thoroughly protect the larvae from parasite 

 attack and to the planting of fruits in which the fly is heavily para- 

 sitized, little of practical value can be expected from the parasites dis- 

 cussed in this paper either in rendering host fruits entirely free from 

 attack or in raising the present quarantine against Hawaiian fruits. 

 In Kona, Hawaii, where the percentage of parasites in coffee cherries 

 has been phenomenally high for two years, it has been high enough 

 merely to render an occasional fruit free from attack. 1 The control 



i The statement of W. M. Giffard that the infestation of coffee cherries during 1914 was at least 50 per 

 cent less than during the previous year, and that in some fields it was difficult to find any great infesta- 

 tion, should be interpreted as referring to the cherries which, although nearly all infested, were infested 

 so late in development that their pulp was little affected when the fruits were picked. 



