A MANOMETER DEVICE FOR AIR THERMOMETERS. 

 By IIahold Edwards. 



Presented by John Trowbridge, June 8, 1904. Received October 10, 1904. 



Five years since an investigation was begun involving the use of an 

 air thermometer. Before the investigation was completed a publication 

 of the Reichanstalt more than covered the field contemplated. The 

 investigation was, therefore, discontinued. In the meantime an air 

 thermometer had been constructed containing some novel features, one 

 of which in particular is perhaps worth recording, — the method of 

 measuring the height of the mercury column. This is a very old prob- 

 lem in connection with both the air thermometer and the normal barom- 

 eter, and it is recognized by every one who has worked with either of 

 these instruments as still far from a satisfactory solution. The best 

 methods used have been those suggested by Pernet, Marek, and Thiesen. 

 These depend on the observations of the direct and reflected images of 

 points or lines through the cylindrical walls of heavy glass tubing, and 

 involve errors due to the displacement and distortion of the images by 

 the non-parellelism of the surface, and by the heterogeneity of the glass. 

 Further, the resolving power of the telescope and the brilliancy of the 

 images observed in these methods are diminished greatly when one 

 brings the mercury surface very near to a fixed glass surface, as is nec- 

 essary at the lower meniscus of a constant-volume gas thermometer. 

 There are besides these many small sources of error familiar to any one 

 who has faced the problem. When the main experiment was abandoned, 

 the work on the air thermometer had been carried far enough to show 

 practically the merits of the device which is here described. 



The general arrangement of the apparatus is shown in Figure 1. 

 A, A are two glass plates containing on their inner surfaces the scales 

 on which the mercury height is to be measured. B, B is the mercury 

 column ; C an air reservoir at the top of the barometer column, exhausted 

 by an air pump in the usual way. D is the tube leading through the 

 wall to the bulb of the air thermometer in the adjacent room. E, E are 

 the devices for determining the position of the surface of the mercury. 



