130 BARRETT—On some novel Thermo-EHlectrie Phenomena. 
the temperature rose to 500°, and after this, only a small change occurred up to 
the highest temperature attainable in the gas furnace employed. The mean 
K. M. F. between 300° and 1000° C., being in round numbers 4000 microvolts, 
and the extreme variations from the mean HE. M.F. throughout this range of 
700° C.,—that is, from a low black heat to a white heat, being less than 170 
microvolts, or about + and — 4 per cent. of the E, M. F. at 300°C., the cooler 
junction being kept at 0° C. It will be noticed there is a slight and curious oscil- 
lation of the E. M. F. about the mean line between 300° and 1100° C., the curve 
cutting the mean E. M. F. at four points, viz., at 310°, 540°, 810°, and 1030° C. 
The couple was next exposed to a very low temperature obtained from solid 
CO;, but though a temperature of — 80° C. was reached (the other junction 
being still kept in melting ice), no other anomalous action occurred, the reverse 
EK. M. F. increasing rapidly as the temperature fell. On again raising the 
temperature to a white heat, the phenomena previously observed were exactly 
reproduced, and continued to be so on repeated heating and cooling. 
In place of nearly pure iron, other substances were tried as the second metal 
in conjunction with this alloy. Ordinary commercial iron wire gave a very similar 
result, the oscillations from the mean E.M. F. being slightly greater. With an 
ordinary mild steel wire as the second metal, the nickel manganese alloy now gave 
a different result; this is shown on the lower curve in the Plate. The E. M.F. 
is less, remains constant only between 400° and 600° C.; then begins to fall and 
continues falling slowly to the highest temperature reached, over 1000° C. 
Platinum, copper, and other metals were also tried in conjunction with this 
alloy, but in no case was there observed the singular constancy of H. M. F. 
through a wide range of temperature which occurred when iron was the second 
metal. With platinum, the HE. M. F. has an opposite sign (to that occurring when 
the alloy is coupled with iron) up to a temperature of 210° C.; inversion takes 
place at this temperature, after which the KH. M.F. rapidly rises with increasing 
temperature in an approximately parabolic curve. 
A small thermo-electric battery was formed of strips of this alloy with strips 
of iron, the strips being insulated by asbestos, and brazed at their junctions; 25 
of these couples give an E. M. F. of 75 of a volt when heated over any flame, the 
cooler junctions being kept in ice-cold water. A convenient standard of HE. M. F. 
might thus be made if the mean of the readings between 800° and 1000° C. are 
taken. Whether repeated heating and cooling of the alloy will affect its KE. M. F. 
I cannot say, but I have not yet observed any injury resulting from this cause. 
We know so little of the whole subject of thermo-electricity, that the ex- 
planation of the remarkable behaviour of this alloy can only be a matter for 
conjecture. Some light may be thrown on it by the results obtained from the 
other alloys of iron, when their thermo-electric behaviour is examined. So far I 
