MR. W. CROOKES OK REPULSION RESULTING FROM RADIATION. 
263 
the glass unchanged, and can therefore do no work on its surface. With the ultra-red 
heat rays, the case is different ; to these glass is almost opaque, and their energy, 
being arrested, is transformed into thermometric heat, or heat of temperature which 
raises the temperature of the glass. But before arriving at the disk of thin glass 
under experiment, the radiation from the candle meets the obstruction offered by the 
front wall of the glass tube in which the disks swing. The heat rays from the candle 
are mostly stopped by this glass and raise its temperature. The warm glass then 
becomes the point whence the lines of molecular pressure radiate ; and it is this mole- 
cular force which repels the plate of thin glass, and not the reaction of molecules 
which are set into activity by a rise of temperature of its own surface. The luminous 
rays from the candle pass unchanged through the front wall of the tube, through the 
thin glass disk, and strike the lampblack with which the hind wall is coated. Here 
they are quenched, and, as I think has been sufficiently proved, they increase the 
temperature of the lampblacked surface and cause molecular pressure in the adjacent 
gas. The disk of thin glass suspended in the centre of the tube is therefore acted 
upon by two opposing forces, one from the front and one from the back wall of 
the tube, and its movement is in accordance with the stronger of the two. The 
front force is due to the direct heat of the candle, whilst the back force is due to 
the light falling on the lampblack and being degraded into heat. Were the heat 
and the light originally equal in amount, the original heat would ultimately largely 
exceed that produced from the degraded light ; but, as is already proved (197), 
much of the action of a candle is due to the ultra-red rays, and it may therefore 
be reasonably supposed that, in comparison to the strong action of the repulsion from 
the front wall of the tube, that from the back of the tube is inappreciable. That such 
an action as is here sought really occurs is abundantly proved further on, but the 
present form of apparatus is not suitable for showing it. 
ACTION OF SCREENS. 
241. When the experiments described in the foregoing paragraphs were first com- 
menced, I used several screens besides water (109, 200, 201), and tabulated the results 
with each kind. They consisted of plates of glass of various colours, and glass cells 
filled with coloured solutions, such as potash bichromate, copper sulphate, iodine 
dissolved in carbon disulphide, &c. But the time required to take a sufficient 
number of observations in order to get a satisfactory mean with each substance 
tried behind each screen, was more than I could give ; I was therefore obliged to 
content myself with a single screen. For several reasons I decided upon water in 
a clear glass cell. First, it is almost perfectly opaque to the invisible heat rays, 
consequently its employment facilitates discrimination between actions due to heat 
and to heat and light combined. Secondly, it is colourless, and, having no selective 
action on any visible ray of light, can be used in conjunction with any coloured 
