384 Miscellaneous Intelligence. 
its length at St. Petersburg according to the relation of the squares 
of the numbers of infinitely small oscillations which the compar- 
calculation gives for the length of the simple seconds’ pendulum 
at St. Petersburg 39°16975 English inches or 441-0319 Paris lines. 
Assuming this length, we have deduced, as the result of our ob- 
servations, the values contained in the first three columns of the 
following table. 
Length of the 
Longitude E. Seconds pendulw Errors. 
Place of observation. Latitude N. from reonwich. in Paris lines. Paris lines. 
Tornea, 1 36 441-2525 +0°0200 
Nicolaistadt, .._... 63 5 33 1 26 26 4411293 —O-0141 
St.Petersburg, .... 59 56 30 a. 441-0319 —0-0017 
Réval 59 26 37 P39 y 441:0190 +0°0033 
Derpet eS 6S aR ae 1 46 54 440°9762 —0-0002 
Jacobstadt, ....._- 56 30 3 1 440°8900 —0°0157 
aha eee tert 5441 2 14112 0°8353 —0°0001 
BM, sd ages 4 62.23 1 40 52 440°7268 —0°0035 
Kréménetz, ....__- 0 6 1 42 54 440°6533 +0-0017 
aménetz-Podolsk,. 48 4 39 1 46 18 440-5844 + 00160 
inchinef; 225 6.1 7 130 1 55 18 440°5278 +0°0030 
WAS accesses 45 20 34 1 55 16 4404479 —0-0071 
To examine the accuracy of each of the results M. Sawitsch has 
compared the length of the pendulum observed at each station, 
with the sovreqonding length obtained from the formule give? 
in hi e residuals are as given in the fifth column of the 
preceding table; the sign + denoting that the observed length 18 
greater than the calculated length. 
The sum of the positive residual errors is +-0°0440, and of the 
negative residual errors —0°0423. age 
Thus the formula agrees well with the equations of condition. 
” 
ff 
above any certain traces of those anomalies and of the local cause? 
which produce them. : 
“In the work of W. Struve on the are of meridian between the 
Danube and the Arctic Sea is a detailed discussion of the latitudes 
of the principal places between the North Cape and the Danube 
The differences in latitude found directly from the astronomica 
observations vary only + 1°75 from those deduced from the ge 
ations. Athough these differences are very much lar 
ie erro, 
