150 THE ROYAL SOCIETY OF CANADA 
resistance in the case of the first two and a decrease of 35% in the last 
one. Undried and dried transformer oil were substituted for machine 
oil and observations made for the metals and alloys as before. The 
same care was exercised in the drying of the transformer oil as in the 
machine oil. The results agree with those obtained when using 
baths of machine oil. The contact pressure in the dried transformer 
oil was in every case less than for the undried oil excepting for con- 
stantan, which appeared to have the same in the undried and dried oil. 
For platinum the value of the pressure in the undried trans- 
former oil was 110, and in the dried 95 mgs. Considerable difficulty 
was experienced in obtaining satisfactory observations for the plati- 
-num contact in transformer oil both undried and dried. The contact 
pressure for silver in the former was 70 mgs; in the latter 62 mgs.; 
for copper 65 and 51 mgs.; for nickel 74 and 71 mgs.; constantan 
97 and 97 mgs.; and for fery-wire 166 and 161 mgs. 
In every case the amount of decrease of the contact pressure 
by drying the machine oil was much greater than in the change pro- 
duced by drying the transformer oil. 
Another point worthy of note is that throughout all the observa- 
tions the greatest contact pressures were those for constantan and fery- 
wire. The chemical compositions of these two substances are similar 
and hence the reason for the corresponding high pressure values. 
In all these investigations well cleaned wires were used. Each 
wire was rubbed vigorously with the finest of emery paper and after- 
wards with a piece of chamois to remove the small particles of the 
loosened material, which cling to the surface of the wire. 
Columns 3, 4, and 6, Table I, give the pressure values for gasoline, 
machine oil and transformer oil. It is shown that the absolute 
pressure depends upon the kind of oil used. 
Columns 5 and 7 give the pressures obtained with the various 
metals and alloys in dried machine oil and dried transformer oil. 
These values indicate that the pressure is less than in the same un- 
dried media. Hence it would seem as if the contact resistance is 
due to the surface conditions of the contact wires. Evidently the 
moisture produces an increase of resistance. 
This substantiates the conclusion arrived at in the article pre- 
viously referred to, that the contact resistance is largely due to a 
cushion effect produced by the formation of a film on the surface 
of the substance, which film in some cases ultimately becomes the 
oxide. It must also be noted that the contact resistance increased 
with time which is undoubtedly the result of a slowly forming surface 
film, the rate of formation being decreased by the exclusion of moisture 
from the contact wires. 
