Prof. W. A. Norton on Terrestrial Magnetism 



209 



years back. The observed declinations for the United States 

 were reduced by Professor Loomis to the epoch of 1837. 



The following table contains the mean annual temperatures of 



the diiferent places in Table I, as computed from Brewster's for- 

 mula, with the values of the distances d and d' used in the calcu- 

 lations of the declinations. The formula is 



T = (*-t) (sin n <Jsin n a')+T 



and with the assumed values of the constants, 



T-(28°^(-19 ))(sin^sin 3T (5 / )-19 =47 (sin^sin^ / )--19 



(8.) 



Table II. 



ans 



j 



an 



» 



gen, 



Place. 



London 

 P 



Mil 



lOottin 

 Berlin, 



^pitzbergen, 

 Moscow, 



Tobolsk, 

 Utga, 



Pekin, 

 Stretensk, 



M. Annual Temp, 

 computed 



Place. 



.lakutsk, 

 Tschernoljes, 



The Azores, 



New York, 



Washington, 



Havana, 



St. Louis, 



Lon.93°VV.,Lat 



Lon.93°W.,Lat 

 Lon.93°W.,Lat 



Sitka, 



M Annual Temp, 

 computedlobservf 



Di/f. 



The observed mean annual temperatures were generally taken 

 from Mahlmann's table, as published in the French edition of 

 Kaemtz's Complete Course of Meteorology. The temperatures 



of the following" places were obtained by estimation from Kaemtz's 

 Chart of Isothermal lines ; viz., Spitzbergen, Tobolsk, Urga, Pe- 

 K»n, Stretensk, Tschernoljes, the Azores, Sitka, and long. 93° W., 

 Jits. 50°, 550^ an j 60 o rp he mean teni peratures for Berlin, Pans, 



London, and Milan, are from observations continued through 

 twenty-five, thirty-three, forty, and seventy years, respectively. 

 * or the other places the observations in no instance extend far- 

 ther back than twenty years. The temperatures for Gottingen, 

 fans, Milan, Berlin, Moscow, and Jakutsk were reduced to the 

 evel of the sea by allowing 1° (centigrade) for from 173 metres 

 19 ° metres of altitude, agreeably to Kaemtz's data. This is a 

 wger allowance than it has been customary to make. According 

 Professor Forry no reduction is necessary for St. Louis. 



*n the calculation of Table II, / was taken in Brewster s for- 

 mula — - - — *' "* 



and 

 cold. 



1 v ery nearly equal to the mean temperature on the equator 



r at a ^„f *l. _ . .£■ *u_ a — ~~;,>r.K> nnl*> nf erreatesl 



temperature 



appears 



^ e both 90° 

 de 



point 



to - This point appears to be situated in Africa a few 



^grees north of the equator. Now Kaemtz gives for the tem- 

 ™ ur e of the interior of Africa in the vicinity of the equator, 

 ■on at an elevation of 300 metres, 29° % Reducing this to the 



■*«*» Ser ies , Vol. IV, No. ll.-Sepl., 1817. 27 



