364 M. Delcros' Barometrical Measurement [Nov. 



These mean limits of disagreement, compared with my 

 measurement of the profile of Jura, appear to me to establish a 

 weight of probability in favour of barometrical measurements. It 

 is true that my measurement is only a peculiar case which cannot 

 constitute a rule. But I can assure the reader that I have cal- 

 culated a great many of my observations, and that they have 

 not only been almost all equal, but often superior in accuracy. 

 This will appear as soon as I make known my other determina- 

 tions. 



The limits 13*38 metres and 41-16 metres given by those 

 measurements in which only a single zenith distance was 

 employed, and in which it was necessary to introduce a hypo- 

 thetical refraction, show evidently that this method is inferior 

 to barometrical measurements, and of course not to be tolerated 

 at present. The coefficient of the refraction is a quantity so 

 variable and so vague, that I cannot express a sufficient degree 

 of astonishment that a constant value is employed for it, appli- 

 cable in all circumstances and seasons. Each mathematician 

 gives his own according to his ideas and his experiments. I have 

 also attempted to form one ; and after having combined a great 

 number of good zenith distances, I have satisfied myself that 

 there is nothing constant in the coefficient but its inconstancy. 

 I have seen it vary between 0*06 and 0*21 of the terrestrial arc. 

 The different times of the day and the changes of the seasons 

 occasion this enormous difference. I believe that sufficient 

 attention has not been paid to the influence of the time of day. 

 In general all the atmospherical phenomena are modified by 

 horary circumstances. The barometer, the thermometer, the 

 hygrometer, the electrometer, the magnetic needle, all point out 

 horary anomalies. The same hours, producing nearly the same 

 effects, ought to occasion the same ordinary terrestrial refraction. 

 This I had an opportunity of recognising on our base of Ensis- 

 heim, in Alsace. This line of 20,000 metres cuts a horizontal 

 plane without obstacle. As a small movement on the surface of 

 the earth prevented us from discerning the signals, we caused a 

 trench to be cut in their direction ; notwithstanding this, I 

 could not observe the southern signal from the north end ; yet it 

 was raised about 20 metres above the ground. An enormous 

 poplar grew near it, and made it more easily be discerned. 

 During the hours about noon I neither could perceive the one 

 nor the other. About three o'clock I began to discover the top 

 of the poplar above the horizon ; some time after, the black ball, 

 constituting the top of the signal, began to appear. I was then 

 able to make my angular observations. I carefully watched this 

 phenomenon, and 1 always saw the signal reach the same posi- 

 tion at the same hours. After three o'clock, the time of the 

 appearance of its summit, it became gradually more and more 

 elevated, showing me first its ball, then its whole pyramid of 

 20 metres of elevation, then its foot, and at last, a little before 



1 



