248 REPORT—1874. 
From the results shown in the Tables it appears that the effect of repeated 
exposure to a red heat for some hours was to cause a considerable permanent 
increase of resistance in pyrometers 404, 411, and 414; while the resistance 
of 445 was almost unaffected by similar treatment, the experiments showing 
in this case a slight fall of resistance. The following Table gives the resist- 
ance of each pyrometer at 10° before and after the series of heatings, the total 
change of resistance undergone by each pyrometer, and the change of tem- 
perature which would produce approximately the observed change of resis- 
tance :— 
Resistance at 10° C. Change of Equivalent 
Pyrometer. resistance at change of 
Before heating. After heating. 10° C. temperature. 
Be ak re) PAT: 8S 
a. a | 9-917 10°749 +0832 430° C, 
INO; EY 2...5 eseaseesees 9-988 11-596 +1:608 +58 
INH 6 take Sos snngoccos 9-920 11-089 +1169 +43 
UNG. 44D an sneusiaaees 10°105 10-059 —0-046 — 15 
The amount of permanent alteration undergone by Nos. 404, 411, and 414 
would probably be considered excessive even in an instrument to be employed 
merely for industrial purposes ; No. 445, on the other hand, though not pos- 
sessing the degree of constancy which would be desirable in a scientific in- 
strument, is probably more constant than any other pyrometer yet devised 
which is capable of supporting equally high temperatures, and would probably 
suffice for most industrial applications. The experiments that have been made 
do not indicate much tendency on the part of the first three pyrometers to 
attain a constant condition : the effect of the later heatings was not decidedly 
less than that of the first. They seem, however, to show that the change 
of resistance is due to the continued action of a high temperature rather than 
to alternations of high and low temperature (compare experiments 5 and7 on 
No. 411, and experiments 9 and 11 on No. 414). Hence it appears probable 
that the change is caused by chemical rather than by physical action ; and it 
has been suggested by Dr. Williamson that it may result from the combined 
action upon the platinum coil of the reducing atmosphere existing inside the 
iron case and the silica of the fire-clay cylinder on which the coil is wound. 
This suggestion is confirmed by the fact, ascertained by Professor Williamson, 
that platinum is readily fused, and apparently becomes alloyed with silicon, 
when heated in a reducing atmosphere in contact with finely divided silica. 
It is also in harmony with the fact that pyrometer 445, in which there was 
no iron in the parts exposed to the greatest heat, did not show a greater change 
than might be attributed to a slight annealing of the wire. Professor Wil- 
liamson proposed, as a means by which the alteration of the platinum might 
probably be prevented, to coat the inside of the iron sheath surrounding the 
coil with oxide of iron, whereby the formation of a reducing atmosphere would 
be made impossible ; and an attempt was made to test the proposal by putting 
some oxide of iron into the sheath of 411; it was, however, thought unde- 
sirable to let the oxide come into contact with the platinum, and the quantity 
which could be introduced without running a risk of its doing so was proba- 
bly too small to produce the intended result; at any rate it did no percep- 
tible good. 
By comparing the results given above, it will be seen that repeated mea- 
surements of the same pyrometer, without intermediate heating, often gave 
almost identical results if they were made within a few days of each other ; 
