146 Level of the Black Sea and tlie Caspian. 



not be reg arded as a definitive result, yet, at all events, by 

 means of this first trigonometrical survey of the Caucasian 

 Isthmus, the fact is established, that the surface of the Caspian 

 Sea is actuallij at a lower letel than that of the Ocean, an opinion 

 which has been lately disputed ; and it is likewise ascertained, 

 that the depression is only about one quarter so great as the older 

 measurements led us to believe. 



Respecting the Barometrical measurement, or the second 

 portion of the operations of this expedition, M. Sawitsch has 

 communicated the following information to Baron A. von 

 Humboldt : — •*' Besides the result of the Trigonometrical sur- 

 vey, we are also in possession of two Barometrical results. The 

 first was obtained by means of observations made at all the in- 

 termediate stations, — stations which were nearly a German 

 mile apart (about 4f English miles). Our instruments were 

 cistern-barometers, like those employed by Parrot. They, as 

 well as the thermometers, were compared daily. At each sta- 

 tion, three observations at least were made between half-past 

 three and half-past six P.M. The difference of level between 

 two adjoining stations generally afforded a satisfactory ap- 

 proximation ; but, during high winds, errors occurred, which 

 for some time were all on one side. These errors of the par- 

 tial measurement were particularly great when a violent gale 

 prevailed, or when there was a regular storm. The definitive 

 result is far removed from the truth, inasmuch as it assigns 

 as the difference of level of the two seas, three hundred English 

 feet, instead of only eighty, which it, in fact, amounts to. 

 The discrepancy between the two results is too great to ad- 

 mit of its being ascribed solely to accidental errors of observa- 

 tions. The sum of these errors cannot amount to more than 

 80 or 40 feet. It may, probably, be rather attributed to at- 

 mospherical influences, which are not always of the nature as- 

 sumed in the usual formulae for the calculation of observations. 

 The second barometrical result is founded on a year's ob- 

 servations made with compared instruments at Taganrog and 

 Astrachan. These give, as the difference of level, nearly one 

 hundred and forty English feet, a result which comes nearer to 

 the truth than the first- At present, an excellent series of ob- 

 servations is in the course of being made at Astrachan and Ni- 

 colajej-, which promises to afford a valuable result." 



