72 hours. As before, there was a strong impression over cell A and none over cell B. The figures were :
Opacity log. = 0'89; Opacity = 7 -71.
These experiments indicate that the electrons from the radio-active agent, chiefly actinium, partake of the properties of a fog or mist of material particles, capable of diffusing away in the free air like odoriferous particles, when not kept in by a thick metal screen.
A further experiment was now tried with the same apparatus, the agent a strongly active radium and barium bromide. This material being self-luminous, a sheet of black paper was placed immediately over it, so that nothing but emanations capable of passing through the opaque paper would be subject to experiment. After 4 hours' expo- sure in total darkness, the film was developed. A good circular patch was obtained over cell A, and a faint diffused darkening showed over the rest of the film, darker at the spot immediately over cell B, fading away at the sides as the distance became greater. That this action was due to the material in the open cell B, and not to general fog over the plate, was seen by the clearness of the film where covered by the lead, and where shadows were thrown by the lead cylinder and pillar.
Circles of the same diameter were drawn round the dark impression over A, and round the darkest part of the impression over cell B. Measurements were taken of different parts of the spaces enclosed in these circles, and the mean of all these came out
Circle over cell A Opacity log. = 0'53 ; Opacity = 3 -39. Circle over cell B Opacity log. = 32 ; Opacity = 2 09.
Katio B/A - 0'62
The experiment was repeated with the addition of a sheet of aluminium, 0'02 mm. thick, under the black paper, the electrons now having to pass through both paper and metal before reaching the film. The exposure was for 6 hours, and the appearance on development was very similar to the last : a dark disc over the protected cell A, and a diffused action over the other part of the film, except in the shadow of the lead supports. Measurements as on the previous occasion gave the following results :
Circle over cell A Opacity log. = 0-78 ; Opacity = 6-03. Circle over cell B Opacity log. = 0-48 ; Opacity = 3-02.
Ratio B/A = 0-5.
A third experiment was tried with the same apparatus, using only the aluminium plate as a screen to cut off the luminous rays. The appearance on development was similar to the others, and the measure- ments of opacities were
