76 TABLE 40 (continued) 



PROBABLE VALUES OF THE GENERAL PHYSICAL CONSTANTS 



quantity can theoretically be determined from any real gas, correcting to 

 reduce to an ideal gas. Actually, only oxygen is used because its atomic 

 weight is 16.000 by definition; there is no error in the resulting value due to 

 error in the atomic weight. As a result of extensive investigations, the cor- 

 rection to change oxygen to an ideal gas is known with considerable accuracy. 

 The I.C.T. gives v n = 22.41 15 x io 3 cm 3 . The H.P. gives 22.4145X10 3 

 cm 3 or R n = 22.4139 liters. The discrepancy must be due to different values of 

 8n(O z ), the normal density of oxygen, or of (1— a), the factor due to the 

 deviation of oxygen from an ideal gas. 1 Thus 



V„ = 32(l-a)/S„(0 2 )=<j32(l-a)/L„(0 2 ) J>I000.027 = i?„ ( IOOO.O27) 



where v n is the normal mole volume in cm 3 , R n the same in liters, 8„(0 2 ) the 

 normal density of 2 , in grams per cm 3 , and L„(0 2 ) the normal density in 

 grams per liter. All these values correspond to normal gravity (g„ = 980.665). 

 It is, however, customary among chemists to express the experimental results 

 in terms of g 45 (980.616). Such values will be denoted by v, 8, L, and R. Thus 



R = M(l-a)/L 

 where M is the molecular weight. 



1 The most general definition of a is (i/pv) d(pv)/d(p), (temp. = constant) ; it 

 measures the change in pv, per unit change in pressure, and has the dimensions of pres- 

 sure -1 . To make the numerical values more definite, it is customary to write a = [i/(/ , f)i] 

 d(pv)/d(p), where (pv)i refers to unit pressure. In investigations on normal density or 

 normal mole volume, it is natural to choose one atmosphere as the unit of pressure. 

 Henning and Heuse use one meter of mercury as the unit of p, and denote a by Kt (see 

 page 85). Since the numerical magnitude of a is proportional to the size of the unit of p, 

 we have Kt = iooa/76. Henning (H.P. 9, 528) uses the symbol Kt, but states that p is 

 measured in atmospheres. 



Within limits of error, the isothermal pv is a linear function of p, for the so-called 

 permanent gases 2 , N 2 , H 2 , etc., for such substances a is independent of p but is a 

 function of temperature, and is more properly written at. The linear extrapolation of 

 pv to p = gives then (pv)o = (1 — a) (pv)\. Now in the limit p = 0, any gas becomes, 

 by definition, an ideal gas. Hence (pv)o is the constant pv of an ideal gas, and (1 — a) 

 is the factor which converts the real (pv)i, (unit pressure) into the ideal (pv)o, both at 

 some definite temperature. (1 — a) is often denoted by (1 + X), and (1 — a) or (1 + X) 

 may be defined as the ratio ( pv) 0/ ( pv) 1. Frequently v is so chosen (in magnitude or unit) 

 that (pv)i is unity, a (or Kt) is then numerically (but not dimensionally) the slope of the 

 pv isothermal (see H.P. 9, 528 and 538). 



Smithsonian Tables 



