98 G. H. KNIBBS. 
throughout, terminated abruptly, and in such cases the resistances 
are greater. Poiseuille in each series, measured the time necessary 
for a certain volume to flow at different pressures. If this volume 
be denoted by Q, and the total efflux time by 7, gin the preceding 
formuls—being the volume per unit of time—is equal to Q/T for 
each pressure. Hence m denoting general value of the coéfiicient, 
i.e. 1°12 theoretically, and supposing that the value P, in (3) may 
be substituted for P’ in (2), we obtain the following expression 
for the viscosity, viz. 
4 
In this formula Q/7' may be substituted for q, that is we may 
write 7’ in the numerator, and @ instead of qg in the denominator, 
and in this way may obtain by transformation an expression for 
what may be called the corrected time of total efflux, viz. 
oe a OTe 
x | eee Pate : 
C=P,T=PT, =Pii mee pe 
A very large number of experiments have been corrected, as bela 
stated, by means of the equivalent of formula (5), Hagenbachs 
value for m however being used. The time 7’, is the total efflas 
time, when the loss of head at the entrance to the tube divided © 
the product of the pressure into the time is negligible, and equa 
tion (6) shews that we may correct either the pressures, OF os 
times. Reverting to Poiseuille’s work, if the pressures in a, series 
of observations made at a uniform temperature with the one tube, 
be multiplied by the corresponding times of efflux, the series © 
products P7' is obtained, which, since the right hand member a 
(6) under the supposed conditions is a constant, differ only by ye 
1 See also formula (40) § 21 hereinafter. . 
2 For the calculation of 79 when Q is unknown see § 21, formula (43). 
a TES rea eres ee 
