July 31, 1884] 



NA TURE 



former case its temperature, i.e. its melting-point, will be 

 situated a few thousandths of a Centigrade higher, and 

 the volume of the ice in the calorimeter will move down- 



t 1 c 



1 :::: 1 fM 



i 



wards on the branch of the curve in the attempt to gain 

 the temperature of the surrounding medium. Then the 

 index will move slowly backwards ; in the other case the 

 co-ordinate of specific volume will move upwards on the 



sloping branch, and the index of the scale-tube will march 

 forwards. These remarks may be sufficient to explain 

 why there has been much dispute about the absolute 



magnitude of the caloric units indicated by the ice calori- 

 meter of Bunsen, and why this instrument is always em- 

 pirically graduated, although its principle ought to allow 

 of absolute measurements. 



II. The amount of heat developed in calorimetric 

 experiments should be directly transformed into work and 

 measuredin absolute ft/zzAfkilogramme-metres]. 1 — In every 

 branch of physical science this manner of measurement is 

 beginning to introduce itself, as, for example, in electricity, 

 magnetism, &c. In thermal determinations it has the great 

 advantage that the mechanic units are 430 times greater 

 than the thermic units, and can be far more accurately 

 determined. 



III. The principle should be applicable to the measure- 

 ment of all kinds of caloric energy : as, specific heat, 

 radiant heat, the heat absorbed or developed by chemical 

 reactions, 6r°c. — It must be remarked that I have hitherto 

 experimented only with radiant heat. An apparatus 

 intended for other kinds of heat is under construction, 



Fig. 4 



but as this will take some time, and as I wish specially to 

 direct my labour to the determination of the sun's 

 radiant heat, I take the opportunity now of calling the 

 attention of physicists to the principle of the method, 

 which I foresee can be developed and varied in a multi- 

 tude of ways. In case the matter should interest anybody 

 sufficiently to make him try it experimentally, I will give 

 some information in order that he may not be detained by 

 those difficulties which have baffled my own efforts during 

 nearly six years. The principle is extremely simple, and 

 will be easily understood by a glance at the diagram 



Let a beam of heat rays fall upon the thi.t glass wall 

 of the reservoir A, which contains dry air. As soon as 

 the thermometer aft indicates that the temperature of the 

 air in A increases, let mercury escape through the stop- 

 cock C. Thereby the air expands, and the heat received 

 by radiation is transformed into mechanic work, as the 

 level of mercury in B sinks under the influence of the 

 expanding force imparted to the air by the radiant heat. 



■ I denote this in the fb:lo\v : ng by the sign M:K°. 



