382 Mr. D. Winstanley on Air-Thermometers. 



long and upright tube. A column of sulphuric acid rests upon 

 the mercury within ; and the summit of the barometric tube is 

 considerably enlarged. The tube connecting the barometer- 

 cistern with the bulb of air is comparatively narrow in its bore. 

 In consequence of this arrangement, the vertical amount of 

 liquid motion is practically confined to the substance of the 

 lighter column, and is several times greater in extent than if 

 one liquid only had been used. We are accordingly enabled 

 in this way to make an air-thermometer, of which the index 

 shall move over pretty nearly its total length for such natural 

 changes in temperature as we meet with in a given place. 

 Such an instrument w r as constructed for me in Paris in 

 1878. It has hung in the Loan Collection of scientific instru- 

 ments at South Kensington Museum for something like a year ; 

 and the accuracy of its indications does not seem impaired. 

 The desirability of using coloured sulphuric acid as the mate- 

 rial of the lighter column will very likely be a matter for dis- 

 pute. I am aware that other experimentalists who have em- 

 ployed it in barometers have found a depression of the column 

 has ensued and a crystalline deposit been left on the mercurial 

 surface. It is for me, however, to speak as I have found. 

 Twelve months ago my instrument, when laid down flat, had 

 a bubble of air in the mercurial limb purposely introduced, and 

 of obviously less volume than a small pin's head. The instru- 

 ment has never been reversed to float it out, and the volume 

 of this air is still capable of the same description as before. 

 This thermometer is four feet or more in length ; and the dia- 

 meter of its index-column (which is cylindrical) is the tenth 

 of an inch or so. Its Fahrenheit degrees are represented by 

 spaces of the third part of an inch ; and it attains an exceed- 

 ingly close approximation to the actual temperature of the air 

 in some few seconds' time ; whilst an alcohol thermometer, 

 moving an equal column through equal spaces for equal num- 

 bers of degrees, and with a bulb similar in shape but propor- 

 tional in size, requires some hours to reach an equally close 

 approximation to the temperature of the air. 



The air-thermometers I have now described have depended 

 for their indications on the movements of a liquid in a 

 tube. I have, however, devised another, in which the move- 

 ment of the tube about the liquid is the method I employ. 

 In this thermometer (fig. 3) the barometric tube is circu- 

 larly curved and mounted concentrically upon a wheel of 

 brass, which is supported in a vertical position by a knife-edge 

 of hardened steel which passes through its geometric centre 

 and rests on agate planes. Adjustments are provided by means 

 of which the centre of gravity of this arrangement can be raised 

 or lowered ; and the mercurial column which is seen extending 



