258 Journal of Agricultural Research voi. v,no.6 



desirable to record at this time the result of the first season's experi- 

 mentation. 



At East Marion, Long Island, N. Y., two fields were selected where 

 during the season of 1914 about 75 per cent of the cucumber vines 

 (Cucumis sativus) had been destroyed by bacterial wilt, as determined by 

 the writer. Here was an excellent environment in which to test the 

 question as to hibernation of the bacteria in soil v. animal carriers. 

 Fifty large frame cages 4 feet square and 3K feet high were constructed. 

 The lower 18 inches of the sides were boarded up, while the covers and the 

 upper 2 feet were inclosed in i8-mesh wire mosquito netting. These 

 bottomless cages were set 15 inches into the soil, leaving 3 inches of the 

 boarded portion above the soil line. The juncture between cover and 

 sides was sealed with cotton and liquid tar, and the cracks between the 

 boards of the basal portion were stuffed with cotton to prevent access of 

 insects. Twenty-three of these cages were set in one of the fields and 

 twenty-seven in the other. In each field the soil in four cages was 

 sterilized by live steam at 75 pounds' pressure for one hour, but this 

 made no difference in the final result. This was done in order to kill 

 any wilt bacteria which might have wintered over in the soil. In each 

 field the cages containing sterilized and unsterilized soil were located at 

 intervals across the field and cucumbers were planted in the usual way 

 in the soil between and within the cages on June 5 and 6. A half-dozen 

 plants were grown in each cage and later on thinned to three or four. 

 After planting, the cages were all sealed with lead seals to preclude acci- 

 dental opening of the covers, and whenever necessary to open the cages 

 for examination they were again sealed in the same manner. By this 

 careful construction and setting of the cages it was thought possible to 

 exclude all of the insects injurious to cucumbers except possibly aphides 

 and flea beetles, some of both of which later on entered some of the cages 

 through the wire netting, but were without eft'ect upon the experiment. 



Field No. i was separated by at least one-half mile, including a quarter- 

 mile depth of woods, from the nearest cultivated cucurbits. It was, in 

 fact, surrounded on three sides by woods and on the fourth side by Long 

 Island Sound. 



Field No. 2 was about one-quarter mile from other cucurbits, but with- 

 out the intervening woodland. 



Plate XXIV, figure i, shows the cages in place in field No. 2; Plate 

 XXIV, figure 2, shows field No. i, with a cage in the foreground lifted, 

 the darker part of the base indicating the depth buried. 



No commercial cucumber fields were planted in either locality until 

 two or more weeks later in the season. 



As soon as the young plants were 2 or 3 inches high and before any 

 wilt had appeared, five or six striped cucumber beetles were introduced 

 into each of 4 cages, 2 in each experimental field. These beetles were 



