188 G. H. KNIBBS. 
well with Poiseuille’s without applying it. This seems hardly a 
satisfactory reason for rejecting an admittedly rational correction: 
the only justification is the one indicated in Section 9 of my 
former paper, see pp. 103-107, viz., that experimentally the 
length-correction appeared to vary and to be either positive 
or negative in sign. The instance selected by Couette from 
Poiseuille’s results and compared with his own, merely chanced 
to agree, and a complete discussion of Poiseuille’s observations 
shewed that the fortuitous agreement was evidentially worthless. 
Aignan’s investigation will perhaps throw some light on this 
point, but it seems that there is no alternative but to follow some 
such method as that previously indicated, or else Couette’s method 
of simultaneous flow through two tubes, using however the more 
rigorous reduction which takes account of the absolute amount 
of the pressures and the ratio of the radii of the tubes, and for 
which I supplied a formula, (43), in my former paper. 
4. Measurements of Viscosity not previously discussed. _ 
Slotte 1883.—Slotte’s 1883 observations! of times of efflux give 
values for approximately every ten degrees of temperature from 
0° to 90° C., and give also a value for 97°C. The relative fluidities 
hereinafter quoted, have been found by correcting his efflux times 
by formula (5) of my former paper (p. 98), m being taken as 1°12. 
In the small interpolation involved in consequence of his observa- 
tions not being made exactly at whole five degrees, second differ- 
ences have been taken into account, as also in the extrapolation 
from 97° to 100° C. Slotte himself had used Hagenbach’s correc- 
tion, i.e. m was taken as 0-79 instead of 1°12. The dimensions 
of his tubes are not known with sufficient accuracy for absolute 
values of the viscosity, but as his observations are discussed only 
for the evaluation of relative fluidities at different temperatures, 
the defect does not prejudice the results. His corrected times of 
efflux, 7, in formula (5) above referred to, range between 1447-2 
eee 
1 Ueber die innere Reibung einiger Lisungen und die Reibungscon- 
stante des Wassers bei verschiedenen betas agrunenanes —Wied. Ann. Bd. 
20, pp. 257 - 267, 1883. 
