68 KENNELLY-SANBORN— INFLUENCE OF [April 24. 



two highest pressures 3.48 and 3.95 megabars. This means that 

 equations (2) and (3) held satisfactorily from i to 2.8 megabars, but 

 did not hold so well outside those limits of pressure. 





Wlt\J'VttocUu CTn~ ^ S<M^rU. 



— j against v, for ^ = 538°. 



In order to eliminate, as far as possible, any disturbing influence 

 on the forced convection in these tests due to the presence of water- 

 vapor in the air contained in the tank, calcium chloride was kept in 

 the tank. The measurements were all made between January 23 and 

 February 17, 191 1, at a time of the year when the air in Cambridge 

 is ordinarily relatively free from moisture. In order to find whether 

 moisture in the air had any considerable effect on the forced con- 

 vection of heat from the test-wire, one test was repeated at one air- 

 pressure (2 megabars) at each of the two temperature-elevations, 

 first with the air after it had been exposed to the calcium chloride, 

 and second with the calcium chloride removed and a dish of water set 

 over night in its place. The actual difference in the humidity of the 

 air in the tank was not measured, but it was supposed that there 



