Art. IX. — Notes on the Barometric Measurement of Heights. 

 By Professor Kernot, M.A., C.E. 



[Read June 13, 1889.] 



I propose at the present time to submit to the Society a 

 few remarks on the above subject, in connection with the 

 very full and interesting paper given by the Vice-President 

 last month. In that paper the avithor expressed considerable 

 doubt as to the accuracy of certain statements that had from 

 time to time been made with reference to the high degree of 

 precision with which differences of level can be obtained by 

 means of ordinary aneroids, in the hands of observers whose 

 time is limited, and who are not provided with the costly 

 appliances to be found at a great national observatory. In 

 this view he is certainly not singular, as several of the most 

 experienced and skilful surveyors of my acquaintance have 

 repeatedly made in my hearing very disparaging statements 

 as to these instruments and their performances. Thus, it 

 will be seen that there has been considerable difference of 

 opinion. Some experienced and reliable surveyors speaking 

 of the instrument as being capable of giving differences of 

 level in average cases within five feet, or even less, of the 

 truth ; while others regard an error of 100 feet as not only 

 possible, but even probable, under ordinary circumstances. 

 To determine which of these two views is correct, is a 

 matter of great importance from an engineering point of 

 view. If the relative levels of points ten or twenty miles 

 distant can be determined with a probable error of not more 

 than ten, or even twenty feet, the location of roads, railways, 

 and schemes of water supply will be greatly facilitated, as a 

 few hours' work of a solitary observer with the aneroid will 

 often serve, instead of several days' labour of a fully-equipped 

 survey party. 



I do not, of course, propose to dispense with accurate 

 levelling in the final or detailed survey of the adopted line. 

 It is in the preliminary investigation and comparison of 

 various possible alternative routes that extreme accuracy is 



