422 Journal of Agricultural Research voi. vi.no. n 



they had begun to disappear from field i. In the variety test the first 

 scattered cases of wilt were observed on July 17, whereas in field i the 

 first cases were noted on July 3. The largest number of cases of wilt in 

 the variety block were found between August 15 and 30, and the striped 

 cucumber beetles were most numerous here during the last part of July. 

 Again, allowing for the necessary time between infection and actual 

 wilting, it will be noted that here also there is a direct relation between 

 the number of wilt infections and the number of beetles present (fig. i). 



The graphs (fig. i) show the daily relation between meteorolo- 

 gical conditions, the number of beetles present, and the number of 

 wilted plants in the three fields from June 10 to August 31. In these 

 graphs there is shown a definite relation between the beetle and the wilt 

 curves, but no relation between the latter and the meteorological curves. 

 The meteorological instruments from which the data were obtained for 

 this graph were kept in a United States Weather Bureau instrument 

 shelter at ground level, so that the environment would correspond as 

 nearly as possible to that of the cucumber plants (PI. LIV, fig. 4). 



Reference should be made to the fact that in taking notes the total 

 number of plants showing bacterial wilt was recorded at each date of 

 observation. This number included not only the new cases but also 

 cases holding over from the preceding observation. Ordinarily the 

 older the plant at the time of infection the longer the interval between 

 infection and death. This explains the apparently too great interval 

 between the maxima of the beetle and wilt curves. If it had been the 

 original intention to represent graphically the relation between the 

 prevalence of the beetles and the occurrence of wilt, the data would 

 have been obtained in a form better suited to this method. It was only 

 after tabulating the results of the field observations that the very striking 

 parallel was noted. Obviously it would be impossible to enumerate 

 absolutely the beetles present in a field; hence, the percentages used in 

 the graphs are based partly on actual counts and partly on careful esti- 

 mates made throughout the season. In the curves, 100 per cent repre- 

 sents the maximum number of striped cucumber beetles present at any 

 one time. 



Attention should be drawn to the fact that although there was a 

 difference of only three days in planting time between field i and field 3, 

 the beetles appeared between two and three weeks earlier in field i, 

 which had been planted to cucumbers the preceding season. This 

 would suggest that these insects hibernated in or near the old cucumber 

 field and that they did not leave this field the following spring as long 

 as young and tender plants remained for them to feed upon. A similar 

 tendency of both striped and twelve-spotted cucumber beetles to hiber- 

 nate in old cucurbit fields was observed by the writers near Congress 

 Heights, D. C. The first frosts occurred in these fields during the first 

 part of October. About the middle of December, 191 5, soil sif tings to 



