282 
MR. J. B. HARRAY OR THE MICRORHEOMETER. 
Salt?. 
Rates. 
Probable error of mean. 
KC1 
129 
0-08 
RaCl 
140-5 
0-14 
kro 3 
126-4 
0-06 
|K 3 S0 4 
143-6 
0-25 
Water 
131-3 
0-06 
A single glance at these numbers is sufficient to show that the crystalline form has 
no recognisable effect on the rate. We find two cubic salts, potassium and sodium 
chloride, acting in opposite directions, the one accelerating the flow by 2'3" and 
the other retarding by 9’2". These it is true are salts of different metals, but 
below we have potassic nitrate and sulphate both trimetric, and yet they diverge 
more widely than the two first, the nitrate accelerating the flow by A'9", and the 
sulphate retarding by 12'3". Solubility, too, does not account for it, because sodic 
chloride is more soluble than potassic nitrate, and yet it retards while the nitre 
accelerates, and potassic sulphate is less soluble than sodic chloride, and retards more 
than it. In fact, some series go one way and some directly opposite; thus in the case 
of barium, strontium, and calcium the latter flows least rapidly, and at the same time 
is most soluble, while in the case of the chloride, bromide, and iodide of potassium 
the first is both least soluble and flows slowest; so that we see that neither the 
crystalline form nor the solubility of a salt materially affect the microrheosis of its 
solution; nor can the specific volume affect it, as we know that isomorphous 
compounds have equal specific volumes. It appeared then that the phenomena of 
microrheosis could only be affected by the mass or energy of the salt in solution. 
It 'may be as well here to explain that as normal solutions are used, mass is 
synonymous with “ atomic weight,” “ equivalent proportion,” or such other term, 
and that when the energy of the salt is spoken of it means the amount of work 
which can be obtained from it by successive combinations till it is degraded to such 
a state that no more work or energy can be obtained from it, which state I describe 
as “ exhausted.” The question is, Does each metal change its value when uniting 
with a acid, or does each metal and each acid radicle have a value of its own 
which is constant? To determine this four compounds were chosen, namely, the 
chlorides and nitrates of potassium and sodium (three of which were already known) 
and their values found— 
Salts. 
Time. 
Probable error of mean. 
Difference H 2 O=0. 
KR0 3 
126-4 
-4-9 \ 
KC1 X 
129-0 
—2-3 \/ + 4 ' 3 
RaCl 
140-5 
+ 9-2 +4-3 
RaR0 3 
137-9 
o-io 
+ 6'6 / 
Water 
131-3 
. . 
0 
When the time taken is less than water I have used a minus sign, although this 
may require to be changed at a future time, but I thought that if we only consider 
