Mr Galbraith's Bai-ometrk Measurements of Heights. 327 



summit, the elevation thus determined falls short of its true 

 height." Dr Anderson of the Academy of Perth, attempts to 

 explain this in the following manner : — " In such a case, it is 

 evident that the intermediate strata between the two stations are 

 placed in circumstances to be powerfully affected by humidity ; 

 of course, the great dilatation which they suffer from the influ- 

 ence of aqueous vapour, tends to increase the altitude of the 

 mercury in the barometer at the upper station ; and by thus 

 bringing the ratio of the pressures nearer to equality, diminishes 

 in a corresponding degree the computed heights by the common 

 formula." 



This seems, at least, to be a rational explanation of the diffi- 

 culty. However, it is proposed to continue our observations, in 

 order to throw some light, if possible, upon such anomalies, 

 arising, in a considerable degree, I am persuaded, from local 

 circumstances, or the partial effects of temperature. 



It has already been observed, that the older made sympie- 

 someter stood at a height somewhat different from the new, and 

 may possibly be ascribed to a gradual deterioration of the instru- 

 ment. No doubt the oil must thicken ; and in the course of a 

 few years its motion must become more tardy. In this case, the 

 only remedy will be a reinoval of the oil, and a re-adjustment of 

 the instrument. But the sympiesometer is not the only instru- 

 ment which suffers deterioration by time. It has been repeat- 

 edly asserted that the mercurial thermometer, and barometer 

 also, become affected by causes for which it is difficult to ac- 

 count, after a lapse of some years. In particular, Mr Daniell, 

 in his Meteorological Essays, appears to prove from numerous 

 registers, that the mercury stands lower in barometers in pro- 

 portion to the age of the Instrument, and proposes a ring of pla- 

 tina, with which the mercury has the property of coming in 

 perfect contact, to prevent the air from insinuating itself be- 

 tween the mercury and the glass, and rising into the vacuum 

 above, thereby shortening the proper height of the mercurial 

 column. It will require some time to verify this satisfactorily 

 perhaps, though the proposer seems, from his own experience, to 

 have sanguine expectations of success. 



It has already been remarked, that the observations by tile 

 barometer should be made simultaneously at the top and hot- 



