492 



Messrs. T. and A. Gray and J. J. Dobbie. [Apr. 24, 



coatings of the flask were connected together until the next reading 

 was about to be taken. For this reading the current was reversed 

 and the deflection taken with its corresponding temperature, and so 

 on throughout the series of observations in any particular case. The 

 electrification was thus reversed between every pair of readings, and 

 lasted in most cases three minutes. The resistances given below are, 

 therefore, those after three minutes' electrification. The constant of 

 the galvanometer for the battery used was determined once or twice 

 in the course of each day's experiments, so that the resistance of the 

 glass might be easily determined in absolute measure from the 

 known thickness and surface of the flask. These were determined by 

 weighing the flask, first empty in air, then with the globular part 

 filled with water, then when immersed in water to a level a little 

 above the former, to allow for the glass wanting on account of the 

 orifice of the neck. The difference between the first and second 

 weighings gave the capacity of the globe, from which its internal 

 diameter on the supposition of sphericity could be calculated. The 

 difference between the second and third weighings gave the volume 

 of the globe from which, as before, the external diameter, and 

 therefore the thickness of the glass, could be deduced. The 

 resistances were then calculated for each specimen and multiplied by 

 the mean surface in square centimetres and divided by the thickness 

 of the flask in centimetres, so as to give the resistance between two 

 opposite sides of a centimetre cube of the glass in question. 



With regard to the results obtained at different temperatures, we 

 have only to say that they confirm the conclusion formerly arrived at, 

 that the conductivity of glass is doubled for every 8' 5 or 9 degrees 

 Centigrade rise of temperature. As, however, the object of the 

 present experiments was mainly to further investigate the relation 

 between resistance and chemical composition, they were not made 

 through any very great range of temperatures. 



In the following table we give the resistance, density, and 

 chemical composition of each of the specimens experimented on and 

 analysed. We give also results for one or two specimens, the 

 electrical resistance and density of which were determined, but which 

 were not analysed. 



The general conclusion from a comparison of these data is, that 

 the specific resistance of the glasses increases with the percentage of 

 oxide of lead, and also with the density. With the exception of the 

 glass marked XX, all the specimens which have been analysed follow 

 the former law, and, with the exception of X and XX, also the latter 

 law, as II, I, and III have practically the same density. But the two 

 flasks X and XX seem anomalous in other respects. They contain for 

 one thing a large quantity of lime as compared with the others, and this, 

 no doubt, tends to diminish their resistance. The density of X 



