350 DR. MEYER WILDERMAN ON THE CHEMICAL STATICS AND DYNAMICS OF 
the plates wires drawn of the same piece of metal from which the plates were made, 
about \ meter long, were fused to serve as leads. I am indebted to Messrs. Matthey 
and Johnson for having kindly prepared the plates for me, and for having supplied 
me with the purest metals it is possible to procure. This is necessary to avoid local 
action between the different metals of the same plates in a solution. The Ag was 
1000/1000, the platinum was of the purest kind, as used for platinum thermometers, 
&c. Special care was taken to avoid thermoelectric currents in the metallic part of 
the circuit. The copper rods or glass tubes to which the plates were fixed serve only 
as supports of the plates, while the current from the plate passes along the thin 
insulated wires drawn of the same piece of metal as the plates, which, coming from 
the ebonite piece, pass in the air behind the asbestos screens of the bath. Thus the 
first contact of two metals in the circuit was removed half a meter from the heated 
plate. These junctions of the two leads in contact with copper were wrapped up 
in ribbons of indiarubber, both then put together and wrapped up with thick 
ribbons of asbestos cloth (subsequently the wires and junctions covered with several 
towels). The terminals of the galvanometer were also wrapped up in indiarubber, 
and then in very thick ribbons of asbestos, and a thick asbestos screen was besides 
placed in front of it, so as to ensure that the incandescent lamp used for photo¬ 
graphing the results should have no effect upon the thermo-couple here. 
Still more interfering in the present research is the thermo E.M. F. created by one 
of the two metal plates in contact with the liquid being heated by light, while the 
other is remaining in the dark, as shown in § 2. This effect can, in the nature of the 
case, never be completely avoided, but it may be very much reduced by the use of a 
bath and stirrer. The quartz vessel was placed into the large bath containing about 
70 litres of water described in my previous communication, its front side being 
8 to 10 centims. away from the quartz window of the bath, so that the light had first 
to pass through a sufficient layer of water. 
Eyen under these conditions it was found that the rise of temperature was still for 
the arc about 0 o, 03 to 0°045 C., when the arc was removed about 27 centims. from 
the quartz vessel. This, as shown before, is in our investigation enough to change 
the obtained results in some cases very essentially. That the effect must he much 
greater when the vessel containing the plates is simply held in the air is self-evident. 
The Preparation of the Plates for the Experiments. 
As already mentioned, the currents obtained with the metal plates were exceedingly 
small, the deflections on Nalder Nl ranging from several millimetres to several centi¬ 
meters. Now 0-000001 of an ampere gives, with my galvanometer Nalder Nl, about 
85 centims. deflection; these 85-centim. deflections are therefore got with about 
0-001 of a volt, the resistance of my galvanometer being 837'4 ohms, i.e., 1 centim. 
deflection of my galvanometer is caused by about 0"000012 of a volt. Now, although 
the plates were prepared of one piece, underwent absolutely the same treatment, and 
