Chemistry and Physics. 219 
be neglected ; with the smallest decrements measured, they only 
amounted to a little over one per cent, and with nearly all the 
measurements employed for calculations much less than one per 
cent, of the total value observed. A substantial simplification of 
the apparatus employed was thus obtained, one vibrating disk 
being sufficient. With pressures of the air varying from 750 to 
380 mms., the absolute coefficient of friction of air was found to 
be 000189. Maxwell found the value ‘000198, Meyer by Max- 
well’s method -000197, Meyer from transpiration experiments 
found -¢00182 and Puluj ‘000185. Calling the coefficient of air 1, 
that of H was -488 and of CO, -806, while Graham’s transpiration 
experiments gave °4855 and °807. Pure aqueous vapor at 21° C. 
and {6 mms. pressure gave 526. To test the theory at low pres- 
sures the chief difficulty consisted in filling the space with pure 
gases. Nothing could be done with caoutchoue connections, 
stand over night. With a pressure of 760 mms. the value of the 
decrement was 0387.—Phil. Mag., \, 53; Proc. Roy. Pruss. 
Acad., 1875, p. 160. BOF. 
9. Conductibility of Heat by Gases—A. Kunpr and E. War- 
ecom 
Sponds to the difference in velocity in the case of friction. The 
€xperiment consisted in the measurement of the rapidity of cooling 
of thermometers of various forms in glass cases of different shapes 
at 0°. At high pressures the effects are masked by air currents, 
but with low pressures this disturbance disappears. Thus with 
4 spherical thermometer the velocity was independent of the pres- 
s.and 1m 
ords an 
heat conduction. Calling that of hydrogen 1, that of air was 
found to be ‘137, and that of CO2, 082 against -141 and "103 cal- 
eulated by Maxwell. The time of cooling of a thermometer in a 
yacuum diminishes if left to itself, in one experiment rape 
