50 TRANSITION RESISTANCE—MACGREGOR. 
As I have already pointed out,* however, it is not clear that im 
the two circuits he used, the number of the times of the current’s 
passing across surfaces of contact was different; and, therefore, the 
equality of their resistance cannot be held to be decisive of the 
absence of transition resistance. 
I had occasion some time ago, in connection with some ex- 
periments on the resistance of solutions of zine sulphate, to de- 
termine whether there was sufficient transition resistance to in- 
terfere with the accuracy of my results. Although the measure- 
ments, which I made for my own satisfaction, apply only to 
one special case, it may be worth while to put them on record. 
The method which I used was essentially the same as that 
of BEETz, but I took precautions which excluded doubt as to the 
number of times the current had to pass from electrode to elec- 
trolyte, and from electrolyte to electrode. I chose as electrolytic 
cell one devised by Prof. Tait, for the absolute measurement of 
the conductivity of saline solutions. It consisted of a box about 
20 cm. long, 10 em. wide and 10 em. deep, made of pieces of 
plate glass cemented together with marine glue. It was divided 
into two nearly equal compartments by a plate of glass cemented 
to its bottom and sides. The two compartments were joined by 
a tube which passed through this plate. On both sides of the 
box, near both ends, narrow strips of glass were cemented to 
form grooves for holding the electrodes ; and the latter were so 
accurately fitted that their position in the box could not vary. 
Thus the column of liquid between the electrodes had a perfectly 
constant leneth. Two similar sets of strips of glass were cement- 
ed on each side of each compartment near those intended for the 
electrodes. These were intended to hold thin plates of zine. 
Thus four such plates could be inserted parallel to the electrodes. 
The current through the cell passed both across these plates and 
around them; but the plates were fitted so accurately that the 
fraction of the current which passed around them must have 
been indefinitely small. The section of the compartments was 
so large relatively to that of the connecting tube, and the plates 
which I used were so thin (about 0.8 mm.) that, though their 
*Proc. R. S. Edin., 1874-5. p. 555: 
