TKANSACTIONS OF SECTION A. 535 



The percentage of tlie total radiation passing through two centimetres of water 

 has been measured at various temperatures. It is shown tliat i'or platinum above 

 1,000° C. this quantity is approximately a linear function of the temperature, the 

 relation being: 



Formula II. 



i! = 872 + 81-G3A 



■where t is the temperature centigrade ; h the percentage of total radiation passing 

 through a two-centimetre Avater-trough having quartz sides tAvo millimetres 

 thick. 



The first formula only applies strictly to a platinum radiator ; the second is at 

 the best a rough approximation. 



They may, however, be of some practical use in many cases where very 

 accurate results are not necessary, or where the temperatures are too high to be 

 directly measured. 



In the second part of the paper tlie absorption curves of a nimiber of substances, 

 including glass, quartz, fluorspar, water, benzol, bisulphide of carbon, &c., are 

 given. 



The curves refer to the radiation emitted by incandescent platinum, the 

 temperature range being from 1,000 to 1,700*^ C. 



For all these bodies the percentage of heat tramsmitted increases with the 

 temperature. 



One exception to the rule is noted — namely, black fluorspar. 



In experimental work it is frequently desirable to maintain the temperature 

 of a radiating body constant; the anomalous absorption of black fluorspar 

 provides a ready means by which this may be done. 



If two thermopiles, one measuring the radiation through, say, glass, the other 

 through black fluorspar, are connected in opposition, there will be a critical 

 temperature at which the resulting galvanometer deflection will be zero. The 

 actual temperature at wliich this will occur will depend on the relative thickness 

 of the plates and on the relative distance of the thermopiles, but for a given 

 disposition of the apparatus it will always be the same. 



_ Some experiments are now being made with a standard of light based in 

 principle on the above results. 



3. Some Observations on Equations of State. By H. H. F. Hvndman. 



1. Equations of state may be obtained by two methods — one by strictly theo- 

 retical deductions from certain given premises checked as much as possible by 

 experimental results, the other frankly empirical and including as many experi- 

 mental results as possible in a systematic manner. 



The only equations considered here will be those which are in accordance with 

 the ' law ' of corresponding states, and they may be generally expressed in the 

 following form : 



f{p,v,t, A,B,C) = 



where A, B, C were independent constants directly expressible in terms of the 

 gas constant II and the three critical constants {p, v, t\. 



