variations of the Magnetic Forces. 331 
nocturnal diminution of temperature may be partially attributable 
to the unequal cooling action of ascending currents of air 
sa law obtains in the calmest nights, andthe assertion so often 
repeated by meteorologists. that in clear calm nights the tempera- 
ture of the soil falls many degrees below that of the air a few feet 
above it, would seem to render such a supposition inadmissible. 
But, whatever view may be taken of the relative part performed 
by the dew in determining the law of the nocturnal loss of tem- 
perature, it is to be observed that it suffices for the explanation 
of the connection .subsisting between this law and. that of the 
nocturnal variations, of the horizontal magnetic intensity ; singe | 
it must be admitted that the tendency of the thermal influence of: . 
same kind with that which actually obtains. However, i 
may possibly also coéperate magnetically with it in determining 
e the law of the nocturnal variations of the horizontal force. 
te _ Having made these ex planations and qualifying statements with 
be “Uy ference to my former memoir, I propose now to show, from 
another point of view, the high probability of the truth of the 
eaPlanation ich | i have, there given of the diurnal variations of 
the hor torce ; and subsequently to discuss the annual vari- 
b* 
short time Reavions to the date of the publication of that 
_ Memoir, [was led, to make a comparison between the curves 
showing the - Gina « variations of the horizontal force, (as given 
in the Report of. the Meteorological and Magnetical Observations, 
made at the » Girard College Observatory, 1840 to 1845,) with the 
furves veal the diurnal variations of the height of the barom- 
peter ; an Rngnient that the following remarkable relations subsist- 
ed bet them... T'hey each have two maxima and two min- 
ima, nate Migynts of the one set nd curves occur at very nearly 
the same hours as the minima of the other. ‘The same relations 
may be observed in the following se sana statement. 
a 
and Barometer, at Philadelphia, for the year 1844. 
i Sa | Horizontal Foree. | Barometer. | 
— Perens | te 
See, peep 5 f M. Midnight. 
NN ae ae eee | 11 p.a. to Midnight. 4a 
It will be seen that i the hours of maxima and minima of the 
barometer differ by not more than one hour from the hours of 
con 4 
tinuing during a portion of the night. ' Although the fact that. 
dew is to produce an inequality of loss of par ere oe a 
Hours of Daily Maxima and Minima of Horizontal Force — 
