1818.] of the Elements of Nitric Acid. 355 



Nitric acid 52-6143 



Potash 47-3856 



100-0000 



Now 47-3856 : 52-6143 :: 6 : 6-662. I consider 6-75 as the 

 true weight of an atom of nitric acid. Now 6*662 approaches 

 very near that weight. 



No person would believe who has not tried it how exceedingly 

 difficult it is to attain absolute precision in chemical experiments. 

 I am persuaded that we can reach that very desirable object in 

 no other way than by taking care that the errors fall upon differ- 

 ent sides, and then taking the mean of a greater number of 

 experiments. If we were in possession of three other sets of 

 experiments made with the same care as those of Berthollet, 

 Wollaston, and my own, upon the analysis of the same salt, I 

 am persuaded that by taking the mean of all the six, we should 

 approach the truth very nearly indeed. The error even from the 

 three does not much exceed one per cent. 



Article V. 



A Barometrical Measurement of the Profile of Mount Jura, on 

 the Line of Geneva, — Lons-le-Saunier, by successive Observa- 

 tions, while corresponding ones were made at Geneva, Stras- 

 burg, and Paris. Executed in 1813 with a Barometer of 

 Fortin, during the Course of a Geodesical Examination. 

 With a Critical Comparison of the Barometrical Measure- 

 ments with those obtained by Zenith Distances. By M. Delcros, 

 Captain ofthe Royal Corps of French Geographical Engineers.* 



(With a Plate.) 



Barometrical measurements would be veiy limited and 

 almost useless if, as some philosophers seem to insinuate, they 

 were only practicable when we have corresponding and nearly 

 vertical observations to compare together. I am of opinion, 

 from my own observations, that this valuable mode of measure- 

 ment may be much further extended, and that it presents an 

 infinite number of applications, which must be useful to the 

 geologist, to the naturalist, to the soldier, and to engineers in 

 general. 



It is of the more importance to point out all possible means of 

 levelling, because hitherto stational topography has hardly 

 entered into the class of operations by which engineers describe 

 l he surface of the earth. We possess excellent plane topogra- 



* Translated from the Eibliotheque Univcrselle, vii. Kit, for March, 1818. 



/. 2 



