fet ON 
From the form for p, in the case of quasi-association, viz. : 
u 2 
al l — — 
RT n—1 | 2 
p= 1 oy) Sl eee 
v 
we derive: 
1 pv a b ús v n—l B x 
PEG ORD)) ob) (ob on GRT ros 
When applying this formula, when large vapour volumes are con- 
cerned we shall no doubt be allowed to put 6 equal to b,, and 
vo—b n 
we 
neglect 4 the second member, and so we shall write this formula 
in the form 
pv Vi; (4 =), hase ) 
— == ib 
RT v sm r v— bg 
@\n—l ve ( fe—1) 
v—bg n v sm 
(3) 
So long as v is large compared with vj we may, of course, put 
VU 
=e 
v—by 
This equation (8) can serve to compute the value of 2 in the 
saturate vapour. 
The value of the first member of (3) for ether has been given 
in the following table according to the mentioned observations. 
WE 10° 20° 30° 40° 50° 60° 
0,0208 0,0219 0,026 0,0835 0,037 0,0375 0,043 
CU 50° JO 2400", iG? EE 
0,052 0,058 0,064 0,066 0,0675 0,07 
HS Ort 140?) 1502) (AGO? OS ae SOE 
0,071 0,07 0,062 0,057 0,045 0,016. 
At 7), this difference, though it need not be equal to O, must yet 
be a small fraction of «7, as I have shown at the conclusion of my 
paper of the preceding month *). These calculated values cannot be 
considered as perfectly accurate; particularly on account of the un- 
certainty in the value of r and of fj. This latter quantity 1 have 
put equal to 7. Probably the value 0,016 at 180° is too low. 
The factor of z in equation (3) is the value of 7a nT 
with a high degree of approximation. At very low temperature the 
1) Proceedings of this meeting p. 1211. 
