METHODS OF MEASUREMENT. 35 



paper could be varied from 25 to 100 cm. per minute. As with the 

 oscillograph, the rate of speed in supplying the paper to the camera was 

 kept as low as was consistent with clear registration, and the time was 

 recorded by means of a Jaquet chronometer. Typical records of the 

 pulse-rate as obtained with the Cambridge string galvanometer, with 

 the modifications noted, are given in A and B in figure 8. 



ELECTRODES. 



In the beginning of the research the subject was connected with the 

 galvanometer by means of zinc rods, wrapped in flannel and moistened 

 with zinc-chloride solution. These he carried in his hands. The 

 difference hi pressure with which the subject at times gripped the elec- 

 trodes caused a varying resistance in the system and led to more or less 

 difficulty. Various other forms of electrodes were tried, but those 

 which were most satisfactory and which were used in practically the 

 whole study consisted of brass disks about 2 cm. in diameter, embedded 

 in kaolin mixed to a paste with dilute zinc-sulphate solution. This 

 paste was plastered on the chest just above and below the nipples and 

 the whole mass covered with a small section of a rubber tennis-ball 

 held in place by strips of surgeon's plaster. The rubber covering 

 tended to prevent the drying out of the moisture in the paste, which 

 would change the conductivity of the system. Over eacji rubber 

 cover was placed a pad of absorbent cotton and one or two large elastic 

 bandages to provide a uniform pressure upon the electrodes. 



The leather belt over the pulleys of the treadmill developed con- 

 siderable static electricity. Furthermore, leakage from the 220-volt 

 system used in driving the treadmill caused considerable difficulty and 

 was a constant source of danger to the delicate string of the galvanome- 

 ter. At first, use was made of the method employed by Benedict and 

 Murschhauser 1 of grounding the treadmill by means of small sections 

 of brass chain trailing over the surface of the leather belt, the chains 

 being connected to a rod and attached to a water-pipe in the laboratory. 

 Later the use of the brass chain was discontinued and it was replaced 

 by a roll of fine brass gauze which bore upon the full width of the 

 leather belt. (See W and Wi, fig. 1, p. 19.) Under thjese conditions 

 it was possible to obtain a record of the pulse-rate with the subject 

 walking and wearing rubber-soled shoes. As at times difficulties 

 would develop unexpectedly, this method was also abandoned, and 

 instead of grounding the treadmill, it was arranged to ground the 

 subject. This was done by a third electrode similar to those already 

 described, which was placed upon the abdominal wall and connected 

 to a water-pipe in the laboratory. (See V, fig. 1, p. 19.) The new 

 method gave excellent results and enabled the subject to walk on the 

 treadmill with ordinary shoes without disturbing the delicate string 

 of the galvanometer. 



Benedict and Murschhauser, Carnegie Inst. Wash. Pub. No. 231, 1915, p. 37. 



