428 Journal of the Asiatic Society of Bengal. [NS., XIV, 
the opposite ends of the torsion arm which was suspended by 
means of a thin quartz fibre within a glass cylinder. When 
light falls on the vanes, the two sides are of course unequally 
heated. But as the air on the two sides is enclosed within 
a glass cylinder, it forms one single system with the glass 
vessel—action and reaction being equal, the radiometer action 
is entirely eliminated. : 
We have found the extra glass cell to be redundant. The 
equally heated by conduction). Thus the radiometer action 18 
entifely eliminated. - 
It will be thus seen that in our arrangement we have fe 
bined the arrangements of Lebedew as well as Hull’s metao®, 
without the additional encumbrance of extra glass cells. 
DESCRIPTION OF THE 
PPARATUS 
The vanes were suspend- 
ed on the opposite arms 
of the torsion balanc : evide 
2.) (m) i 
meter mirror | placed at 
right angles to the plane of 
the vanes, with a small 
from the excursion of a spot of light reflected from ae 
usual lamp and scale arrangement. The dimension 
