2 35 
influence in the solar rays. 
In whatever way we may attempt to account for the ter- 
minal arc of vibration, when a needle is screened from the 
sun, exceeding that when it vibrates exposed to the rays, 
(which excess I have called the terminal excess) the result 
here obtained with the glass needle is certainly extraordinary : 
as compared with that obtained with the magnetised needle, 
it shows decidedly that the solar rays exert a peculiar influ- 
ence on a magnetised needle during its vibration : considered 
separately, either that glass is magnetic, which is scarcely 
probable, since its magnetism is not detected by rotating a 
magnet under it, or that the rays produce an effect distinct 
and independent of their magnetical effect. 
These observations were made consecutively without even 
moving the compass-box ; but it is scarcely possible, especi- 
ally at the present season of the year, to make such a series 
under precisely the same circumstances of temperature of the 
air and of solar heat, by which alone we can in this case 
judge of the intensity of the action of the sun. The change 
which took place during the observations was probably not 
quite so great as I have registered, since in the first observa- 
tion with the magnetised needle the glass had been closed 
over the thermometer for some time before I registered its 
height ; and in the first observations with the other needles 
it had been open and in the shade, for the preceding observa- 
tions, until very nearly that time ; so that 100,99 and 98 
would probably indicate more correctly the comparative ele- 
vations of the thermometer in the first observations with the 
different needles than 100,96 and 93, which I have regis- 
tered. But even taking the latter, the small difference 5 0 
would produce little difference in the results, and is much 
