48 BULLETIN 491, U. S. DEPARTMENT OF AGRICULTURE. 



cooperate ( 1 ) in the destruction of the eggs and larvae by submergence 

 in water and (2) in the destruction of the adults by spraying, the 

 value of spraying with a poisoned bait, and of covering the young 

 fruits, would be enhanced to a point where either might be sufficiently 

 effective to be recommended as satisf actorj^ ; but so long as the culti- 

 vation of host plants is largely in the hands of uneducated orientals 

 who are not amenable to instruction, no relief can be expected. 



Spraying. — Since adult melon flies do not begin ovipositing until 

 from 10 to 14 days after emergence, depending upon the tempera- 

 ture, but feed continuously throughout this period, it is evident that 

 any spray that will kill them before they begin ovipositing is 

 valuable. Severin (45) and Marsh (27) have shown that poisoned- 

 bait sprays are capable of killing adult flies in large numbers, 

 as has already been demonstrated by Mally and Bezzi in the case 

 of the Mediterranean fruit fly. But good results promised by 

 the work of Mally and Bezzi were found by Severin, Marsh, and 

 the writers in Hawaii to be offset by adverse conditions, such as 

 frequent showers and lack of concerted action by those growing host 

 fruits. Marsh, who sprayed beds of cucumbers with a solution made 

 of 1 quart of molasses, one-fourth ounce of Paris green, and 1J gal- 

 lons of water, daily from September 9 to October 14, obtained nega- 

 tive results. This was true also when arsenate of lead was used 

 instead of Paris green. Severin found that sprays containing a solu- 

 ble poison, such as potassium arsenate or sodium arsenite, killed off 

 enormous numbers of adults quickly but " burned the foliage, and 

 can not be advocated." He was unable by spraying to protect young 

 watermelon plants, because he could not control the outside sources 

 for adults and because of the frequent showers. 



The writers have obtained negative results in their work with the 



poisoned spray: 



Lead arsenate (paste) 5 ounces. 



Brown sugar 2+ pounds. 



Water 5 gallons. 



About 2 acres of Chinese melons and cucumbers in a field fairly 

 well isolated, which from the Hawaiian standpoint means that no 

 host fruits were growing within 500 yards, were sprayed on May 21, 

 26, and 28, June 1, 4, 8, 14, and 23, 1915, during typical summer 

 weather. At the beginning of the experiments the young fruits 

 were badly stung. Every fruit under 2 inches long was infested 

 unless it had been properly covered, and adult flies were very numer- 

 ous. The spray was applied at the rate of about 30 gallons to the 

 acre by means of a knapsack sprayer. The abundance of living adults 

 was so noticeably lessened after the third application that even the 

 Chinese gardeners noted the fact, and for the first time during the 

 experiment showed a willingness to cooperate. The interval be- 



