35 



tions have been derived from the elaborate discussion by 

 Prof. E. Plantamour of the long series of thermometrical and 

 barometrical observations made at Geneva and the St. Ber- 

 nard, as a means of ascertaining altitudes, in the " M6moires 

 de la Societe de Physique de Geneve." They are calculated 

 for use in connection with the class of hypsometric tables 

 founded upon the barometric formula of La Place, to which 

 Dr. Guyot's tables belong. 



The total correction thus applied for the influence of the 

 hour of the day, contains two main corrections, one for the 

 difference in amount and in time, above and below, in the 

 daily tide of the barometer, the so-called horary (hourly) 

 variation ; and another, by far the largest, for the error in the 

 true mean temperature and moisture of the layer of air 

 between the upper and lower barometer, as given by the 

 observation of the instruments at both stations. As the last 

 depends upon conditions of the atmosphere extremely vari- 

 able, which cannot be reduced to a formula, it is evident that 

 the amount of the total correction has to be modified accord- 

 ing to the various states of the atmosphere existing at the 

 time of the observation. When the sky is clouded, or during 

 a fog, and with a gentle wind mixing the layers of the atmo- 

 sphere, the average temperature of the air, given by the arith- 

 metical mean of the upper and lower thermometers, will be 

 much nearer the true one than when the air is calm, the sky 

 clear, and the insolation and radiation excessive. In the latter 

 case the layers nearer the ground, in which the instruments 

 are necessarily placed, show abnormal temperatures, which, 

 according to the hour of the day at which they are taken, are 

 either above or below the mean temperature of the whole 

 mass of the air. 



We have, therefore, somewhat modified the amount of the 

 correction when it seemed advisable to do so, after a careful 

 consideration of the atmospheric circumstances indicated in 

 the accompanying meteorological record. 



We feel confident that by this method, the greater part, if not 

 the whole, of the influence of the hour of the day is corrected. 



The elements of the corrections having been derived from 

 observations made between 1000 and 9000 feet, it remains to 

 be seen whether they are equally applicable to higher alti- 

 tudes. This matter remains for future investigation. Thus 



