1883.] Mr McConnel, On Dark Rings in Quartz. 53 



arrangement as that for the determination of the tension of 

 aqueous vapour in the air, by the chemical method of absolution 

 of the moisture by desiccating substances, with the addition of an 

 apparatus for previously saturating the air. 



The ordinary form of the apparatus is modified by avoiding as 

 far as possible india-rubber connections, using glass tubes and 

 mercury joints for making communication between the different 

 parts. The saturating apparatus is similar to that used by Regnault. 



The desiccating substances must be either sulphuric acid or 

 phosphoric anhydride; chloride of calcium does not completely 

 dry the air passed over it. 



With such an apparatus, which is easily constructed, corrections 

 for a thermometer enclosed in the saturating vessel can be 

 determined which, for the thermometers experimented on, agree 

 with the Kew corrections to within a tenth of a degree. 



o 



(3) Measurement of the Dark Rings in Quartz. By J. C. 

 McConnel, B.A. 



MacCullagh has explained the peculiar optical behaviour of 

 quartz, by introducing into the equations of motion of light terms 

 involving differential coefficients of the third order. These equa- 

 tions contain in addition to the two principal wave velocities only 

 one arbitrary constant. So it becomes of interest to see if all the 

 facts are really accurately contained in this simple theory. I 

 believe the only published observations on the subject are those of 

 Jamin in 1850. He made three series of experiments on plates of 

 quartz cut perpendicular to the axis, using sodium light. In each 

 case he employed parallel rays. In the first series he used a 

 Babinet's compensator to observe the nature of the light originally 

 plane polarised, which had passed through a plate of quartz. By this 

 means he obtained values of the retardation of one ray in the 

 quartz relative to the other, and of the ratio of the axes of the 

 ellipses of polarisation of both rays. In the second series he 

 inclined a plate of quartz between two crossed Nicols and observed 

 the angles when the transmitted light was completely quenched. 

 This would happen whenever the relative retardation was an 

 exact number of wave lengths. In the third series he examined 

 the nature of the transmitted light at those points in the preceding- 

 experiment where it reached a maximum. He found that his 

 observations confirmed the theory within the limits of experimental 

 error. But these limits were not very close, as the average differ- 

 ences between the observed and the calculated values were about 

 ^th of the quantities to be determined and there were several 

 wide discrepancies ; one or two amounting to ^th. 



I thought that I would try if a more accurate determination of 



