Oscillations of the Horizontal Needle. 19 
rays ; and their views have received much confirmation from the 
many important inquiries of Mr Curisrre, and which have at va- 
rious times appeared in the Transactions of the Royal Society. 
These researches are of much consequence to experiments with 
the oscillating needle. If a magnet be caused to vibrate in 
the shade in air, and subsequently when exposed to the rays of 
the sun, the time in which the arc of vibration becomes reduced 
to a given point will in the sunshine be considerably less, whilst 
the rate of vibration seems upon the whole to be somewhat in- 
creased. 
In briefly detailing the following experiments made with va- 
rious bars, both in the shade and in sunshine, I would by no 
means pretend to have decided the question, whether the solar 
ray has really magnetic properties or not. My object is merely to 
shew, that certain differences observed to arise in the ares and 
times in vibration, under a given solar influence, may be caused 
to vanish by placing the needle in an exhausted receiver. 
32. On exposing an oscillating magnetic bar to the influence of 
a bright sunshine, so as to compare the phenomena incidental to 
the state of vibration with those previously observed in the 
shade, all the effects so accurately noticed by Mr Curistt1e be- 
came strikingly apparent ; but whether the bar be caused to vi- 
brate in a dense, or in a rare medium, the experiment, under 
an air-tight glass, is quite unmanageable for any accurate pur- 
pose. The sources of inconvenience appear to be, Ist, an al- 
teration in the rare or dense medium surrounding the bar, to- 
gether with a constant circulation of currents; 2d, a condensa- 
tion of vapour over the interior of the glass; 3d, various mecha- 
nical and other changes in the conditions of the suspended bar 
itself. These circumstances rendered the experiments in a mid- 
day sun so very precarious, that I was obliged to abandon them. 
33. (d.) With a view of avoiding the above mentioned sources 
of inconvenience, I eventually conducted the experiments in 
the following manner. Having observed the oscillations in the 
c2 
