as applied to the measurement of Heights. SS5 



for the measurement of heights, unless in peculiarly favourable 

 circumstances. * 



The experiments from which I expect to deduce these re- 

 sults were made by me so far back as 1825 : -|- having been 

 led by some previous observations to doubt the speedy accom- 

 modation of the instrument to the condition of the air, as stated 

 by Mr Adie in his small work which accompanies the sym- 

 piesometer, I resolved to institute a series of observations 

 on the subject upon the same spot of which I had already ob- 

 tained the approximate height, and by taking the indications 

 at short successive intervals, to discover in how long a time the 

 sympiesometer might in all cases be considered to have taken 

 its final level, which is the most important practical question 

 in the use of an instrument. It was with much vexation that 

 I found, after many trials, that this time seemed quite indefi- 

 nite, and often appeared as if it would never arrive, as even 

 after hanging an hour, the oscillation continued. I threw the 

 observations aside as hopeless ; and it was only at a later 

 period that I saw some chance of drawing inferences from them 

 which might explain the anomaly they disclose, and now, after 

 nearly four years, I give them to the world in hopes of excit- 

 ing farther inquiry on the subject, and to suggest, or at least 

 point out, the mode of discovering a cure for the defect. 



It will be proper, first, to explain what I conceive to be the 

 great fault of the instrument, and then endeavour to substan- 

 tiate it by the detail of my observations. It will readily be 

 admitted that the cylindrical receptacle at the top of the instru- 

 ment, which we may suppose on an average to be 2 inches long 

 t and I in diameter, which is filled with the most diffuse known 

 fluid, hydrogen gas, must be incomparably sooner affected by 

 any change of temperature than the large bulb of a thermo- 

 meter, intended to have degrees of great size, and which ex- 



• Besides these objections, a very important one has been established by 

 a very complete set of experiments, recorded in the Edinburgh JEnci/clopw- 

 dia. Art. Meteorology , p. 173, that a gradual absorption of the gas by the 

 oil takes place, raising the indications of the instrument. 



t I must here acknowledge the important assistance which I received in 

 ihe prosecution of these experiments from ray brother Mr C. Forbes. 



i 



