52 
Proceedings of Royal Society of Edinburgh. [sess. 
‘ a ’ is small and also x , equation (ii) is really a particular case 
of equation (i) ; for we may put (i) in the form 
, , . z 2 log e 2 A 
Y) x = 1 + x\og e A + 
91 I 
£ 3 log e 3 A 
"* V. + 
or, putting log e A = a 
a 2 x 2 a s x? 
rj v = 1 4- ax + -yy + -yr + • • • . (hi). 
Considering aqueous solutions, we may (roughly) divide the 
dissolved substances into electrolytes and non-electrolytes. In 
the former class substances are known, e.g. potassium chloride, 
which do not follow the above formula (iii), hut possess what may 
be called a ‘ negative ’ viscosity. Thus the viscosity of J normal 
potassium chloride is less than that of water. Up to the present 
no non-electrolyte has been found to show this ‘ negative ’ 
viscosity. In the paper mentioned above, Rudorf drew attention 
to the fact that carbamide in dilute aqueous solution shows a 
‘ negative ’ viscosity. I have repeated these measurements, and 
have also made determinations of the viscosity of acetamide in 
solution. These substances show a normal behaviour in their 
depression of the freezing-point.* 
Carbamide (Urea). 
Concentration. 
Vi 
V2 
A 
T V mol. 
1-005 
... 1-005 ... 
•o 
i „ 
1012 
... 1-011 ... 
- -001 
h 
mol. 
1-024 
... 1-022 ... 
- -002 
1-015 
... 1-045 ... 
2 „ 
1-089 
... 1-092 ... 
+ •003 
Acetamide. 
Concentration. 
Vi 
Vz 
A 
| mol. 
1-013 ... 
1-014 ... 
+ •001 
i >> 
1*028 ... 
1-028 ... 
i „ 
1 "057 ... 
1-057 ... 
mol. 
1-117 ... 
1-118 ... 
+ -ooi 
o 
“9) 
1-250 ... 
1-250 ... 
^ is the viscosity determined experimentally ; rj 2 that calculated 
from equation (iii). A is difference of the calculated value from 
that observed ; (a) in the case of carbamide being taken as *044, 
* Zeits'chrift fur pliysikal. Chemie, 2, 491 (1889). 
