U. S. DEPARTMENT OF AGRICULTURE 



The lack of uniformity in the yields of the control plats A and C in 

 the 1907 experiments (Table 1) is such that no great dependence can 

 be placed in these results. It is significant, however, that in only one 

 of the 10 trials recorded did the treated plat show any evidence of a 

 substantial increase in yield when compared with the mean of the 

 control plats. 



Experiments in 1908. — In the 1908 trials the wires were run di- 

 rectly over the treated rows and kept at a height of 6 to 18 inches 

 above the plants by means of adjustable brackets on which the 

 insulators were mounted. The control rows ran parallel to the 

 treated ones at a distance of 6V2 feet and were separated from them 

 by intermediate guard rows. 



In one part of the plat the wires over the plants were charged 

 positively to about 50,000 volts from 4 p. m. to 7 a. m. each day, 955 

 hours in all. In the other part of the plat the wires were charged 

 and discharged rapidly by connecting them to one terminal of the 

 secondary of an induction coil, the other terminal being grounded. 

 In this case the potential rose to about 20,000 volts and then dis- 

 charged suddenly through a small spark gap between the wires and 

 the ground. 



The treatment first described is similar to that employed by Lem- 

 strom and believed by him to result in increased yields. In these 

 experiments, however, neither treatment gave any evidence of in- 

 creased growth. The detailed yields consequently are not of special 

 interest. 



ELECTROCULTURAL FIELD EXPERIMENTS WITH GRAINS 



In selecting a location for the electrocultural field experiments near 

 Washington, three conditions were sought: (1) A uniform soil, (2) 

 available electric power, and (3) accessibility from the laboratory in 

 Washington, since the equipment had to be visited daily during the 

 experimental season. Soil uniformity is particularly difficult to find 

 in the environs of Washington, and the Arlington Experiment Farm 

 forms no exception in this respect. It seemed to be the best avail- 

 able location, however, and portions of sections A, B, and E were 

 made available for the experiments, which were carried on from 1911 

 to 1918. Sections A ana B proved very disappointing with-regard 

 to their uniformity, and the most reliable results were obtained in 

 section E. These experiments will be first described. 



The Lodge-Newman apparatus used in the experiments from 1912 

 to 1915, inclusive, was designed in England primarily for electro- 

 cultural work and consists essentially of a 110- volt induction coil, 

 operated by a mercury interrupter, and a rectifier. Five Lodge 

 valves 4 designed to rectify the high-tension alternating current were 

 placed in series with the network, thus allowing only the positive 

 impulses from the secondary of the coil to reach the network {33). 

 The negative pole was grounded. Two balls 25 millimeters in diam- 

 eter, one of which was grounded and the other connected to the net- 

 work, were used to determine the potential, assuming a breakdown 

 gradient of 3,000 volts per millimeter. 



Systematic measurements of the current from the network were 

 not made, but the current could be determined approximately from 

 the potential of the network and the known power characteristics of 



* For a description of the valves, see Lodge, O. (84). 



