182 VITALITY AND EFFICIENCY WITH RESTRICTED DIET. 



on the rubber. It also decreases the resistance which must be over- 

 come by the arm A in passing under. The sleeve tends to roll on the 

 rubber band; the position of wear is thus continually shifted. This 

 form of catch is without objectionable sound and, in fact, is almost 

 noiseless. 



The handle H, shown incomplete in figure 44, extends 75 cm. beyond 

 the periphery of the disk and, passing through a table above the pendu- 

 lum, is arranged for slow movement with a rack-and-pinion device as 

 shown in figure 47 (see R and P). By this means the separation of the 

 switches S and S' (figure 44) may be continuously and accurately 

 varied if desired. In the reduced-diet investigation the separation 

 of the switches S and *S' was always the same, 5°, representing a shock 

 duration of about 0.0035 second, and only the voltage was changed. 

 It is of course evident that the voltage might be held constant and the 

 threshold determined by changing the duration of the shock by shifting 

 the position of >S' with the slow-motion control of figure 47. Unpub- 

 lished experiments have shown that a threshold may be determined in 

 this way. 



The electrodes which were used for this measurement recommend 

 themselves for several reasons. In the first place, it is of prime impor- 

 tance in making sensory measurements with the fingers as receptors 

 that the hand should be in a normal and comfortable position. The 

 arm should be supported, the hand relaxed, the fingers should not be 

 required to span from one vessel of solution to another in such a way 

 that the muscles are tense and must be frequently rested. The form 

 of the electrodes employed in the present research and the convenience 

 of their application may be seen in figure 48. A two-compartment 

 glass vessel (pickle-dish), each compartment 8 by 9 cm. in area and 2.5 

 cm. deep, was suitably supported at a level conveniently below that of 

 the armrest. Soft pads were provided which the subject could arrange 

 under the palm of the hand according to his own desire and comfort. 

 Two fingers could very easily be placed in the salt solution, one on 

 either side of the narrow glass partition. Two porous clay cups, one 

 located in each compartment, contained a saturated solution of zinc 

 sulphate. Amalgamated zinc rods were also placed in the porous 

 cups. Thus the electrodes were comfortable and non-polarizable. 

 The liquid could be brought to body temperature and easily controlled 

 by the small electric heater which was located below the electrodes. 

 Another point in favor of these electrodes, involving considerable 

 economy of time, was that the area of each compartment was suffi- 

 ciently large in comparison to the volume of the finger-tips immersed 

 in them to make it unnecessary to adjust the height of the solution for 

 each subject. The depth of immersion employed was 2 cm. 



Figure 48, which has just been described, shows the fingers of the 

 subject in position for stimulation. In figure 49 the hands of the 



