4*4 



Prof. A. Schuster on the Scale- Value of 



Table II. 



Date. 



Height 

 of Baro- 

 meter. 



Freezing- 

 point be- 



Freezing- 

 point after 



Interval. 



Depres- 

 sion of 

 zero. 



fore ex- 

 periment. 



experi- 

 ment. 



a. 



b. 



April 7, 1892. 

 May 11, „ 

 May 30, „ 

 June 22, 1894. 



7591 

 769-1 

 759-7 

 766-0 



1504 



16-58 

 4-97 



9-88 



14-35 



16-04 



4-46 



9-08 



690-79 

 689-63 

 690-97 

 690-69 



691-48 

 69017 

 691-48 

 691-49 



•69 

 •54 

 •51 



•80 



Mean -64 



The agreement between the experiments, excepting the 

 second, is better than could have been expected, and accident 

 must have played some part in giving these practically 

 identical numbers. The first and third determinations were 

 made by myself, the fourth by Mr. J. R. Ashworth, the 

 second by another observer, and there must be some error in 

 it which could not afterwards be traced. The intervals are 

 given in 2 columns — (a) is the interval calculated on the old 

 method of taking the freezing-point first, while (b) is that 

 now generally employed, the reading at the boiling-point 

 being compared with the reading at the freezing-point 

 taken immediately afterwards. The difference between the 

 number so obtained and Joule's interval (690*95) gives the 

 depression of zero as *53, agreeing fairly well with that found 

 directly by the above experiments. The interval (a) is also 

 seen to agree with Joule's value. The time the thermometer 

 was kept exposed to the temperature of boiling water varied 

 from a quarter of an hour to several hours. In the first 

 three experiments the depression seemed to increase with the 

 time of exposure, but in the last experiment that time was 

 only about fifteen minutes, i. e. shorter than in the other 

 cases ; the large depression may be due to a diminution of the 

 column by distillation. 



It appears, therefore, that the depression of the zero is less 

 than C *1. It approaches that observed in Jena or French 

 hard-glass thermometers, and is considerably smaller than 

 that found with modern English glass. 



The Thermometer A. — This thermometer does not include 

 the boiling-point, but its freezing-point is of interest, as it is 



