689 
of Edinburgh, Session 1879-80. 
able to add to the general experiments with the differential and 
ordinary telephone above described. In particular, I have repeated 
the experiments of Grant in several striking forms, and made a 
variety of others of a similar nature. I have gone into the 
mathematical theory of these results, and, I think, succeeded in 
explaining the Curious changes in the quality and intensity of the 
sounds observed. Among the results, I should desire particularly 
to draw attention to the theory of the striking alterations produced 
by condensers in the jpitch y or, more correctly speaking, quality of 
telephone sounds. 
2. On the Determination of the Specific Heat of Saline Solu- 
tions. By Thomas Gray, B.Sc., Demonstrator in Physics, 
and Instructor in Telegraphy, Imperial College of Engi- 
neering, Tokio, Japan. Communicated by Professor Tait. 
The object of the present paper is to describe the results obtained 
and the mode of experimenting adopted in some determinations 
which I have made of the specific heats of solutions of salts. These 
experiments form part of a series which I am at present carrying, 
out on the physical changes produced when salts are dissolved in 
different amounts of their solvents. From such an investigation I 
believe much information may be gained regarding the nature of 
solution. 
The method of experimenting adopted in the experiments de- 
scribed below was that of mixtures ; but as regards the mode in 
which the exact amount of heat added to the solution was measured, 
it differed from any process with which I am acquainted. This 
peculiarity consisted in using as heater a thin glass bottle of about 
50 cubic centimetres capacity, and furnished with a long glass 
neck, just wide enough to allow an ordinary mercury-in-glass thermo- 
meter to pass through. This bottle was nearly filled with mercury, 
in which was immersed the bulb of a sensitive thermometer, and 
thus the temperature of the mercury in the bottle could be read 
off at any instant. The graduation of this thermometer was to 
fifths of a degree centigrade, and had been compared with the Kew 
standards. The distance between two consecutive divisions of its 
scale was about one millimetre. 
